You are on page 1of 161

UNIVERSITY OF SOUTH AUSTRALIA

School of Advanced Manufacturing and Mechanical


Engineering

Bachelor of Engineering

Final Year Project

Wave Energy Potential for Kangaroo Island

Tom Winkler

I.D. No: 100005269


Supervisor: Dr Brian Kirke

2005

1
Abstract

Renewable energy technologies offer power for a sustainable future and there are two
key factors driving their development. Firstly is the growing awareness of the need to
reduce the destructive effects of conventional energy use on humanity and the
environment. The second factor is to reduce the heavy reliance society has on burning
limited fossil fuel reserves, oil in particular. Of the “new” renewables, Wind energy is
leading the charge by being the most economically viable form of renewable energy
technology at present. However, other forms of renewable energy are emerging
including harnessing the power of the ocean in the form of wave energy generation.
Ocean wave energy is an abundant and highly concentrated form of renewable energy
that up until very recently has been virtually unexploited. Wave energy now looks set
to become an important addition to the renewable energy mix.

Local renewable energy developer Wind Prospect is exploring the use of a specific
wave energy generation technology, the Pelamis P-750 Wave Energy Converter,
developed by Ocean Power Delivery Ltd in the UK. The Pelamis is viewed by Wind
Prospect as the leading contender in commercially available wave energy generation
technology and having the potential to become a valuable supplement to wind power
for supplying Australia with renewable electricity and possibly desalinated water.

The original purpose of this project was to assess the feasibility of installing a ‘wave
farm’ to contribute to the electricity demands of Kangaroo Island, South Australia.
Wind Prospect required an investigation into the use of Pelamis wave energy
converters in a specific offshore location. After it was discovered that the feasibility
for Kangaroo Island was not as good as anticipated, the scope broadened significantly.
This project explores the available wave energy resources and bathymetry of the
entire coastline of southern Australia. As a result it has identified Australia’s best
locations for offshore wave energy projects. Portland in Victoria was identified as the
most promising and its feasibility is outlined in a case study.

2
Research was undertaken to establish the nature of ocean wave energy and the long
term wave climatology arriving at southern Australia. Over 10 years worth of wave
data was analysed and combined with Pelamis specification data to model and predict
the power output of the Pelamis at specific sites. Investigation is made into issues
associated with grid connection, electricity demand and local Pelamis manufacture
and assembly. A brief life cycle analysis is undertaken to estimate the embodied
energy of a Pelamis and other environmental impacts.

Wind Prospect has recently become interested in the potential for renewable energy to
desalinate water on an industrial scale. This is an emerging area of research and Wind
Prospect required a conceptual investigation into the feasibility of offshore sea water
desalination using the Pelamis. A number of different configurations are proposed for
using the mechanical pumping power of the Pelamis to drive saline water through
Reverse Osmosis units within its structure.

3
Disclaimer

I hereby declare that this thesis is my own work and contains no material, which has
been accepted for the award of any degree or diploma from any tertiary institution. To
the best of my knowledge and belief, this thesis contains no material previously
written or published by another person, except where due reference is made in the
text.

Signed: …………………..

Tom Winkler
7th November 2005

4
Acknowledgements

I would like to thank my project supervisor Dr. Brian Kirke for all the guidance, input
and help throughout the project. You have helped me gain a much better
understanding of renewable energy technology and improved my project management
skills. Our sometimes overly lengthy discussions, which moved off topic, were not at
all a waste of time in my opinion. I would also like to thank Michael Vawser,
Managing Director of Wind Prospect and Andrew Dickson, Development Manager.
Thank you for developing and supporting the project and for having the time for
meetings, emails and for providing general help and sourcing information. Special
thanks to Andrew, who provided great help during the latter 2/3 of the project, for the
endless search for wave data, and for proof reading the entire thesis only a couple of
days before the submission date.

I also sincerely acknowledge the assistance provided by the following individual and
organizations:
• Vincenzo Bellini (Vinni) from Ocean Power Delivery for answering my
questions on his trip to Adelaide. It’s unfortunate that the trip to Edinburgh
didn’t eventuate.
• Edward Mackay and Helen Plowman at Ocean Prospect, Bristol, for the
teleconference and emails, even if I could not get you data and the MCP
was not able to be used in my project.
• Mike Lewis, General Manager of Air Ride Wind, for giving us a tour of
the wind turbine tower plant, it a very interesting and useful experience.
• Paul Driver and the team from ETSA Utilities for the meeting regarding
the connection enquiry for Kangaroo Island.
• Osmoflow for data on reverse osmosis units.
• Powercor for offering assistance with the Portland connection enquiry.

I would also like to mention the support of my parents, family, friends (the dudes I
live with who had to put up with ****) without which would have made life much
harder. Also my experience at the Whyalla steelworks which helped me gain an
appreciation for energy use in industry, an appreciation of embodied energy and a
greater respect for the impacts of energy use on the environment.

5
Contents

Abstract ................................................................................................................... 2
Disclaimer .............................................................................................................. 4
Acknowledgements............................................................................................... 5
Contents.................................................................................................................. 6
List of Tables........................................................................................................ 12
1 Project Background and Significance ....................................................... 13
1.1 Aim of the Project.................................................................................. 13

............................................................................................................................ 14
1.3 Project Scope.......................................................................................... 18
1.4 Identifying Audience............................................................................ 20
1.5 The Energy Industry and Renewables ................................................ 21
1.5.1 Climate Change and Environmental Concerns ................................. 21
1.5.2 Energy Demands, Global and Local.................................................. 24
1.5.3 South Australia’s Energy Snapshot.................................................. 25
2 Literature Review and Project Plan........................................................... 27
2.1 Renewable Energy Technologies......................................................... 27

6
2.1.1 Common Issues with Generating Renewable Power ......................... 27
2.1.2 Overview of Renewable Energy Technologies ................................... 30
2.1.2 Energy Storage and Transportation Technology............................... 35
2.2 The Power of Sea Waves ...................................................................... 40
2.2.1 Oceanic Wave Formation and Climatology ...................................... 40
2.2.2 Wave Energy Characteristics ........................................................... 44
2.3 Wave Energy Converters...................................................................... 49
2.3.1 Study of Different Designs............................................................... 49
2.3.2 Projects at Advanced Development Stage......................................... 52
2.4 The Pelamis Wave Energy Converter ................................................. 54
2.4.1 Design Concept and Features ........................................................... 54
2.4.2 Performance and Specification.......................................................... 55
2.5 Project Discussion and Methodology.................................................. 57
3 Southern Australian Wave Energy Resources ............................................ 59
3.1 Australia's Long Term Wave Climate ................................................. 59
3.2 Study of Waves Arriving at Cape Du Couedic .................................. 62
3.2.1 Data Description.............................................................................. 62
3.2.2 Statistical Analysis and Wave Climate Trends................................. 64
3.2.3 Wind to Wave Correlations and Directional Climate........................ 73
3.3 Study of Waves Arriving at Cape Sorell ............................................. 75
3.3.1 Data Description.............................................................................. 75
3.3.2 Statistical Analysis and Seasonal Climate Trends ............................ 76
3.4 Survey of Australian Coastline Bathymetry ....................................... 79
3.5 Measure Correlate Predict Model for Waves ..................................... 82
3.6 Conclusions ........................................................................................... 83
4 Case Studies for Potential Wave Farms .................................................... 85
4.1 Kangaroo Island Wave Farm Study .................................................... 85
4.1.1 Summary of KI Energy Review ........................................................ 86
4.1.2 Grid Connection and Location Bathymetry ...................................... 87
4.1.3 Pelamis Performance Model at Cape Du Couedic ............................. 89
4.1.4 Recommendations ............................................................................ 92
4.2 Portland Wave Farm Study.................................................................. 93
4.2.1 Location Bathymetry and Grid Connection ...................................... 93
4.2.2 Resource and Performance Prediction .............................................. 96

7
4.2.3 Recommendations ............................................................................ 97
4.3 Other Potential Pelamis Projects.......................................................... 98
4.3.1 Cape Sorell and other Tasmanian locations ...................................... 98
4.3.2 West Coast of Eyre Peninsula .........................................................100
5 Pelamis Production and Development ....................................................102
5.1 Local Pelamis Construction.................................................................102
5.2 Developments in Desalination............................................................103
5.2.1 Overview of Reverse Osmosis Technology.......................................105
5.2.2 Pelamis Design Integration.............................................................109
5.2.3 Offshore Desalination Plant Concepts.............................................112
6 Life Cycle Analysis.....................................................................................114
6.1 Embodied Energy for a Locally Manufactured Pelamis...................114
6.2 Energy Balance .....................................................................................120
6.3 Conclusion ............................................................................................122
7 Conclusions and Recommended Further Works ..............................................123
References ...........................................................................................................126
Appendix A .........................................................................................................133
Ocean Power Delivery ......................................................................................133
Pelamis P-750 Brochure...................................................................................133
Appendix B .........................................................................................................134
Statistics of World Energy Use.......................................................................134
Appendix C .........................................................................................................137
Summary of Wave Energy Information Resources ....................................137
Companies and other Non-literature Resources...........................................137
Literature Resources Explored .......................................................................139
Appendix D .........................................................................................................143
Descriptive Coastal Wave Photographs .......................................................143
Appendix E..........................................................................................................145
ERA-40 Wave Atlas Plots .................................................................................145
Appendix F..........................................................................................................148
Extracts from BOM Cape Du Couedic Buoy Report...................................148
Data Buoy Cooperation Panel : Scientific and Technical Workshop XVII ..............................................149

The South Australian Wave Rider Buoy and ......................................................149


Some Preliminary Comparisons of Wind and Wave Data. ................................149
Andrew Watson ...........................................................................149

8
1. Introduction ...............................................................................................150
2. Wave Rider Buoy Description.....................................................................151
Appendix G .........................................................................................................152
Description of Wave Data Fields ....................................................................152
Appendix H .........................................................................................................153
Bathymetric Image of Australia and the Southern Ocean.........................153
Appendix I ...........................................................................................................154
Geography of Cape Du Couedic to Cape Sorell..........................................154
Appendix J ..........................................................................................................155
Comparison of Embodied Energy Values ....................................................155
Appendix K .........................................................................................................156
Environmental Impact Assessment - Generic Scoping Study .................156
Appendix L..........................................................................................................157
Industrial Experience Report ............................................................................157

9
List of Figures

Chapter 1
Figure 1.1 Artist’s impression for a 40 Pelamis, 30 MW ‘wave farm’ 13

Chapter 2
Figure 2.1 The Vestas V120 31
Figure 2.2 Concentrating solar thermal collector types 32
Figure 2.3 Current Hydrogen consumption 36
Figure 2.4 Illustration of wave fetch area to Cape Sorell 40
Figure 2.5 Annual average wave power in kW/m of crest length for various 41
locations around the globe
Figure 2.6 Monthly mean 500 hPa wind speed (m/s) and direction, Southern 42
Hemisphere, July 2005
Figure 2.7 Gravity Waves 43
Figure 2.8 An idealized sinusoidal ocean wave 44
Figure 2.9 Plot for equation (2.4) empirically relating wind speed to wave 44
power
Figure 2.10 Illustration of approximate particle orbits 45
Figure 2.11 A typical record from a Wave Rider Buoy 46
Figure 2.12 Evolution of UK Wave Power Devices 48
Figure 2.13 Configurations of Wave Energy Converters 49
Figure 2.14 Energetech’s WEB Design 52
Figure 2.15 Pelamis Performance Matrix 54
Figure 2.16 Relative efficiency of a Pelamis 55

Chapter 3
Figure 3.1 Global ERA 40 Plots for yearly mean Hs and Tz 59
Figure 3.2 Sample of wave rider buoy data 61
Figure 3.3 Series of plots from combined monthly mean values 64
Figure 3.4 Plot of monthly means for all 3 ½ years of Hs and Tz data 65
Figure 3.5 Daily mean values for winter ’02 (left) and summer ’01/’02 (right) 65
Figure 3.6 2003 Histograms for Hs 67
Figure 3.7 2003 Histograms for Tz 67
Figure 3.8 2003 Histograms for Power flux 67
Figure 3.9 Overall Histograms for Hs 69
Figure 3.10 Overall Histograms for Tz 69
Figure 3.11 Overall Histograms for Power flux 69
Figure 3.12 Scatter Diagram for all year 2003 recordings 70
Figure 3.13 24 hour distributions from mean hour values 71
Figure 3.14 Hs and wind speed against time 73
Figure 3.15 Hs and wind direction against time 73
Figure 3.16 Hs (m) against wind speed and direction 73
Figure 3.17 Combined monthly mean plots and comparison 76
Figure 3.18 Monthly means over the 7 years of data 77

10
Chapter 4
Figure 4.1 Kangaroo Island’s Electricity grid network 84
Figure 4.2 Coastal bathymetry of the region south of American River 86
Figure 4.3 Pelamis performance curves for various Tz values 88
Figure 4.4 Near linear property of Tz curves 89
Figure 4.5 Mean monthly Pelamis output over the whole 3 ½ years 89
Figure 4.6 Annual Comparison of Mean Pelamis Outputs 90
Figure 4.7 Coastal bathymetry surrounding Portland with depth in metres 93
Figure 4.8 Cape Duquesne Location 93
Figure 4.9 Electricity Grid Surrounding Portland 94
Figure 4.10 Comparison of Pelamis performance from combined monthly 96
means
Figure 4.11 Satellite image of Cape Sorell 98
Figure 4.12 Satellite image of Cape Du Carnot 99

Chapter 5
Figure 5.1 Basic components of a RO plant 105
Figure 5.2 Hollow Fibre RO membrane configuration 106
Figure 5.3 The Aquadam Concept 112

Chapter 6
Figure 6.1 Energy used by Bluescope to make steel 115

11
List of Tables

Chapter 1
Table 1.1 The World’s Proven Fossil Fuel Resources 23

Chapter 2
Table 2.1 Annually Averaged Extractable Energy Flux 33
Table 2.2 Energy Densities of Various Fuels 37

Chapter 3
Table 3.1 Total number of recordings per month 2001-2003 62
Table 3.2 Summary of Mean Values at CDC 63
Table 3.3 Total number of recordings per month 1998-2004 75
Table 3.4 Summary of Mean Values at Cape Sorell 75

Chapter 4
Table 4.1 Summary of Mean Capacity Factors for Pelamis at CDC 90

12
1 Project Background and Significance
1.1 Aim of the Project
Wind Prospect is in the business of expanding the renewable energy industry both in
the UK and in Australia. A recent endeavor by the company is to explore the use of a
specific wave energy generation technology, the Pelamis P-750 Wave Energy
Converter, developed by Ocean Power Delivery Ltd. (OPD). The Pelamis is viewed
by Wind Prospect as the leading contender in commercially available wave energy
generation technology and having the potential to become a valuable supplement to
wind power for supplying Australia with renewable electricity and possibly
desalinated water. The aim of this project is to asses the feasibility of developing
projects using the Pelamis and the wave energy resource so abundant along the
southern coasts of Australia.

Wind Prospect decided that Kangaroo Island has particularly good potential for off-
shore wave energy production, a potential worthy of greater and more detailed
investigation. Initially the scope of this project focused on assessing the feasibility of
a 4 unit ‘wave farm’ (see Figure 1.1) demonstration off the shores of Kangaroo Island.
Since then the scope has been modified and broadened to include other potential sites
and to focus investigation into other aspects of local wave energy development. The
case study approach was adopted and Kangaroo Island is the first to be investigated.

This project is industry-based and is primarily intended to cater for the interests of
Wind Prospect. Ultimately Wind Prospect has the goal of convincing outside
investors and manufacturers to join in supporting development of real wave energy
projects. This report was to be completed with sufficient detail and supporting
evidence to greatly assist Wind Prospect in reaching their goal.

This project is the first investigation into wave energy undertaken by Wind Prospect
in Australia and many aspects of the research to be undertaken are completely new.
The purpose of this report is to not only outline the feasibility of producing potential
wave farms but to provide a knowledge resource for Wind Prospect. Each of the
components included in the scope (section 1.3), and in particular the study into wave

13
energy resources, will become a valuable reference for current and future work by
Wind Prospect in the field of wave energy.

Due to the heavy research nature of this project compared to a conventional


mechanical engineering final year thesis, the scope has been subject to continuous
change and review. As new developments occur, the focus has moved toward new
targets and priorities. Incorporated into the aim of this project is for research to be
initially relatively flexible and for continuous consultation with Wind Prospect to be
sought in the interests of best project management. In this way, the project will be of
far greater value to industry.

Figure 1.1 Artist’s impression of a 40 Pelamis, 30MW ‘wave farm’ (OPD 2005)

More images of and information on the Pelamis can be found in the Ocean Power
Delivery brochure, APPENDIX A and in section 2.4.

14
1.2 Project Introduction and Background

The project is a new collaboration between the Sustainable Energy Centre (SEC) of
the University of South Australia (UniSA) and Wind Prospect Pty. Ltd. It is a final
year mechanical engineering student project and principally aims to be highly
successful in meeting the assessment criteria set by the university. The final draft
project thesis was to be completed by the 7th of November 2005.

The official industry partner of this project is Wind Prospect and the work aims to
best meet their requirements. However, it is also a channel for research in the field of
renewable energy technology. In the first few months, the author spent considerable
time researching and becoming educated about renewable energy worldwide. This
was done in order to obtain a clearer understanding of the industry and was essential
to visualising how the core scope of the project’s work fits into the bigger picture.
Given that this was a feasibility study and the hardware supplier has already been
tentatively selected, the literature review also investigates in further detail (i) energy
issues in a global context, (ii) available wave energy converter designs and
technologies.

The concept of this project and UniSA’s involvement began through discussions
between Dr. Brian Kirke of the SEC and Wind Prospect during September and
November 2004. The key facilitator of this project is the SEC of UniSA located
within the school of Advanced Manufacturing and Mechanical Engineering. The SEC
is dedicated to researching, developing and promoting environmentally friendly and
sustainable technologies. Among its attributes, the SEC offers academic support and
resources towards nominated projects offered to final year mechanical engineering
students.

Manufacturers and developers must make a profit to stay in business. Creating


prototypes such as the Pelamis is an expensive process and is often not cost
competitive. Wave energy converters, as with all forms of newly explored
technologies, need a learning curve before the technology becomes viable. The case
for wave energy converters is discussed in more detail in section 2.3. Once promising
technologies evolve to the stage of near commercialization, there is a need to

15
convince governments and venture capitalists that the technology is worth investing in
real projects. Again, this is a costly and time consuming process. This is why Wind
Prospect has sought the resources of the SEC and a University student project
dedicated to providing this service at a minimal cost.

Wind Prospect is a wind farm developer that was established in the UK in 1988. The
Australian division, Wind Prospect Pty Ltd., is currently based at Christies Beach,
South Australia. Wind Prospect claims an unrivalled success rate in securing planning
consents and is a leading wind farm developer in Australia. They undertake all aspects
of development including feasibility studies, design, construction and operation. With
the aim of maximizing the cost-effectiveness of every project, all ventures of Wind
Prospect remain independent of manufacturers. Wind Prospect has been involved with
over 1,300 MW of wind energy developments worldwide, including 700 MW in
Australia. Using experience gained with onshore projects, Wind Prospect established
SeaScape Energy Ltd. to develop offshore wind energy projects in the UK. It must be
noted that Wind Prospect and all other renewable energy developers depend greatly
on government inscentives to make their projects financially viable.

Ocean Prospect is a recent venture established by Wind Prospect in the UK to develop


marine renewable energy projects. Wave energy is seen by Wind Prospect as an
important part of the future mix of renewable energies. Ocean Prospect is currently
focused on working with OPD to develop projects using the Pelamis off the shores of
the UK. Following the initiative of Ocean Prospect in the UK, Wind Prospect now
wants to explore wave energy opportunities in Australia.

Ocean Power Delivery Ltd. (OPD) is the designer and manufacturer of the Pelamis P-
750 Wave Energy Converter and is based in Edinburgh, Scotland. They employ a
wide range of engineering expertise and have 40-50 cumulative years of staff
expertise in wave energy. They are a world leader in ocean wave energy analysis and
modelling. OPD have produced and tested a full scale prototype Pelamis, claimed to
be the most advanced wave energy converter available. It is a result of 6 years of
development, modelling and testing and OPD claim at least an 18 month lead over
rival companies. The Pelamis technology is proven and ready for commercialization
and OPD are sourcing developments across the globe. OPD are a manufacturer and

16
rely on developers such as Wind Prospect to develop projects using their machines.
The Pelamis design is protected by OPD under a group of patents and patents
pending. A significant component of the cost of a Pelamis is the steel, structural
manufacturing and labour costs. OPD prefer that as much of the manufacture of these
components is outsourced to local specialized manufacturers, such as those that build
steel wind turbine towers. Only certain groups of internal components are to be
manufactured by OPD and then shipped to the location of assembly. Once a Pelamis
is completely assembled, it is then thoroughly inspected by OPD before deployment.

Air-Ride Wind Pty. Ltd. is a business dedicated to manufacturing wind towers and
was established in 2003. It is a South Australian owned and operated company based
at Kilburn North in Adelaide. They are involved with all assembly and manufacturing
from supplied steel plate, through to fully assembled and painted wind tower
components with internal structuring that are ready for final on site assembly. They
are able to manufacture tapered welded steel tubes up to 4m diameter that can be
transported using rail and road freight. Air-Ride work closely with wind farm
developers including Wind Prospect. They also work closely with industry leading
European wind turbine designers to ensure that all standards and specifications are
met. All steel plates are currently supplied by Bluescope Steel, based at Port Kembla,
NSW. Air-Ride prides themselves on quality welding procedures and inspection and
on meeting target delivery schedules. It is through companies like Air-Ride that
renewable energy projects can be seen positively contributing to our local SA
economy.

17
1.3 Project Scope

As stated in the project aim, the scope of this project has been under continuous
change and review. This scope is effectively the last major revision which occurred on
the 4th August. Information in this scope refers only to work done beyond the
literature review, Chapter 2.

This project deals only with analyzing the feasibility of the Pelamis designed by OPD,
not with any other wave energy converter. No attempt will be made to redesign the
Pelamis, its manufacturing and construction procedures, its component specification
or its installation and mooring design. The only exception is for the desalination study
(section 5.2) where a redesign of the existing internal layout of the power modules
will be investigated to incorporate reverse osmosis units.

For the study of wave energy resources (Chapter 3), only the southern coast of
Australia is investigated. This is done in the order of SA, VIC, TAS and WA. The
study should result in identifying Australia’s best wave energy resource locations
ready for further development. Wave data from Cape Du Couedic is analysed using
statistical methods only; no advanced mathematical wave modelling is attempted. The
task of creating the “measure correlate predict” (MCP) mathematical model (see
section 3.4) for wave resources in Australia is left to Ocean Prospect in the UK.

Case studies of potential wave energy projects (Chapter 4) will contain all work
undertaken on the Kangaroo Island study. This then forms a template model for later
case studies, which are less in-depth. The case studies use the information from
Chapter 3 to predict wave energy levels and then investigate aspects of project
development including: localized bathymetry, demand load, grid connections,
political and integration issues such as docking and maintenance. The Pelamis output
at the Cape Du Couedic buoy is also modelled to determine a likely Capacity Factor
and output trends. The result is a basic design of a ‘wave farm’ to suit each case
study.

18
Chapter 5 deals with Pelamis production and development explores three areas:
construction, component sourcing and desalination. The local feasibility of
construction of a Pelamis is investigated although no planning for production is
undertaken, as that work will be left to the manufacturer(s) (Air-ride, Bluescope etc.)
All internal components that are to be sourced locally are identified and listed. Effort
will go towards finding suppliers willing to provide components to the designed
specifications. The study into desalination, as mentioned above, is a design
investigation to lay the path for potential future research by Wind Prospect and OPD.

A basic lifecycle analysis is undertaken for a locally manufactured Pelamis based on


one of the case studies (either KI or Portland). A calculation is made, stating all the
assumptions, to determine a realistic figure for the embodied energy of a Pelamis unit
and the consequent energy and CO2 payback periods. Comparisons are then made
with the lifecycle analysis of a Vestas wind turbine. An environmental impact
assessment is then produced for the case studies using the OPD EIA as a template.

19
1.4 Identifying Audience
Aside from UniSA (contents subject to the confidentiality agreement – possibly
modified), the full project thesis will be read by and is directed to the following list of
companies or groups in order of importance:

• Wind Prospect Pty. Ltd. (Australia) and Wind Prospect (UK) including all
their venture businesses

Subject to confidentiality agreement with Wind Prospect, a special version(s) of this


report may be compiled (even if by Wind Prospect) for viewing by:
• Ocean Power Delivery Ltd.
• Various grant agencies
• Air-Ride Pty. Ltd.
• ETSA Utilities
• KIDB – Kangaroo Island Development Board
• ESIPC – Electricity Supply Industry Planning Council
• SENRAC – South Australian Energy Research Advisory Committee
• ESCOSA – Essential Services Commission of South Australia
• BOM – Bureau of Meterology
• Separate investors linked with Wind Prospect
• Various other relevant government departments and authorities

A version of this report may eventually be made freely available - ie. published on a
website subject to the removal of copyrighted content and confidential information.

20
1.5 The Energy Industry and Renewables

In its simplest definition, renewable energy is any form of energy that can be taken
from a renewable source. All of the energy on Earth and most of its matter originated
from the Sun – our solar system’s giant thermonuclear power plant. The only possible
exception to this is matter from outside our solar system, maybe distant meteorites.
The sun is considered to be the only true renewable energy source even though its
energy is not limitless because the sun won’t last forever. For this reason sometimes
the term solar energy is used to cover all the forms of renewable energy including
wind, wave, hydro and biomass, because they did originate from the sun. Geothermal
energy is usually considered to be renewable although some individual sources have
been weakened from heavy exploitation. Tidal energy is renewable but is generated
by the gravitational pull of the moon as well as the sun, so could be termed “lunar
energy”. All fossil fuels such as coal, oil and natural gas are the heavily compressed
remains of hundreds of millions of years of biomass. The energy needed to grow this
biomass and store chemical energy in their long hydrocarbon chains was all provided
by the sun. However, fossil fuel sources are definitely finite and are not renewable.
For an investigation into the global renewable energy sources see section 2.1.2.

A lot of study and debate has gone into the idea of using renewable energy over the
last few decades. There are a lot of factors involved but there are two main reasons
why renewable energy has become important. Firstly there is a growing need to
reduce the destructive effects of energy use on human and environmental health. The
second is to reduce the heavy reliance society has on burning fossil fuel resources, oil
in particular. An appreciation of these issues must be gained in order to understand
the appeal behind using energy generated from ocean waves rather than by current
conventional methods.

1.5.1 Climate Change and Environmental Concerns

The expansion of the human species has brought about massive changes to our planet.
The global human population is now over 6.476 billion inhabitants (IPC 2005), a
growth of about 10 times the population existing just 300 years ago and the number is
projected to climb even further. In meeting the demands of our population, the earth’s
ecology has endured the effects of widespread habitat destruction and land clearance,

21
pollution and climate change. Great concern is growing over the state of our ancient
natural ecosystems and their suffering biodiversity. Of particular threat are the
imminent effects of climate change due to global warming.

Global warming or the ‘greenhouse effect’ is an important phenomenon and has been
required for sustaining life on earth for hundreds of millions of years. The earth’s
atmosphere is a very complex system and there are many factors that contribute to the
atmosphere’s ability to trap the suns heat. Contrary to popular understanding, water
vapour is by far the most powerful contributor to the greenhouse effect. Water vapour
along with aerosols, clouds and ice reflectivity can further amplify any small
increases in global warming (NAS 2005). A great deal of research is yet to be done in
order to gain a comprehensive yet irrefutable understanding of the process and status
of global warming. The fact is, our planet is experiencing a growth in atmospheric
temperature and this research is a critical tool for addressing solutions to this problem.

Numerous scientific authorities support evidence that suggests recent human activities
are contributing to global warming in a harmful way. After water vapour, carbon
dioxide (CO2) is the gas next most responsible for trapping the sun’s heat in our
atmosphere, and it can remain there for very long periods. Studies show (IEA 2002)
that since 1750 the atmospheric concentration of CO2 has grown 31 percent. Increases
in global temperatures have already been detected and ‘during this century, the global
average surface temperature is projected to increase by 1.4 to 5.8 °C’ (IEA 2002,
p.19). The magnitude of this temperature change could be compared to that during the
last Ice Age over 20,000 years ago. The projected increase in global temperature
would have devastating results. Sea levels will rise due to melting ice at the poles.
Severe weather including droughts and floods will increase due to altered weather
patterns and ocean currents. Also there would be shifts in climate zones of hundreds
of kilometres, which could cause mass extinctions and damage vast ecosystems. The
threat posed to humanity’s future is enormous.

The other significant contribution to global warming is the effect of wholesale land
clearance for agriculture and for wood products. Forests are still being cleared at the
fastest rate in history, much faster than they can be replaced. Trees and all other plants
take up CO2 and place it into the soil and use the carbon in new growth. About 2.6

22
tons of CO2 is consumed by an acre of trees in a year (ISA 2005). Even if CO2
emissions are dramatically reduced, its concentration in the atmosphere will not
decrease without the work of forests. Additionally, great expanses of cleared land for
agriculture are heating the air far more than forests do. This allows higher
atmospheric water vapour concentrations, further amplifying global warming (NAS
2005). Careful, directed and realistic efforts to sequester CO2 in its natural stores is an
important factor in combating global warming, but alone it is not a solution.

Burning fossil fuels creates CO2 and this is widely considered a major cause of
accelerated global warming. Other emissions are also generated from burning fossil
fuels that are harmful to the environment and human health. This includes other
greenhouse gases but also pollutants such as: carbon monoxide, sulfur oxides,
nitrogen oxides, ozone and fine particulate matter. Air pollution has become a major
issue in industrialised countries and a lot of regulations and spending programmes are
directed towards cleaner air. Most forms of renewable energy do not produce any
emissions during operation.

In order to reduce the severity of climate change, an enormous effort will need to be
made by all the nations of the world to reduce CO2 concentrations in our atmosphere.
A big part of this will be to use energy more efficiently and to find and use
alternatives to burning fossil fuels, such as renewable energy technologies. In 1997 a
summit was held at Kyoto in Japan, where a group of nations adopted a protocol
committed to reducing greenhouse gas emissions of industrialised nations. The aim
was to reduce greenhouse emissions to 1990 levels by the year 2000. It was only a
start, because a reduction of 60-80 percent below those levels is required. Today,
atmospheric CO2 levels are rising faster than ever. There has never been a time in the
history of earth when the surface ecology has been so dramatically altered to serve
entirely for the demands of a single species. As a population we must reduce our
‘ecological footprint’ and become a global culture focused on ecological
sustainability. ‘In our hands now lies not only our own future, but that of all other
living creatures with whom with we share the earth’ (Attenborough 1979, p.308).

23
1.5.2 Energy Demands, Global and Local

Our current system of economics promotes meeting the unlimited demands of the
population by exploiting finite resources. Since the dawn of the industrial age over
150 years ago, the pace of technological development has accelerated to this day
largely fuelled by fossil fuelled energy. As a result of this our demand for resources
and energy is growing endlessly. Importantly, the world’s current energy consumption
rate greatly exceeds what is required to sustain a good quality of human life for all
populations. Fossil energy comes into all aspects of modern human life including
goods, services, comfort, transport and industry. The International Energy Agency has
published a report (IEA 2004) outlining statistics for the world’s energy usage. Some
pages from this report can be found in APPENDIX A showing the world’s energy
supply and electricity generation grouped by fuel. From this data it is very easy to see
modern society’s reliance on fossil fuels (coal, oil and natural gas) as an energy
source. Looking at electricity alone, total of 16.1 percent of all consumed energy was
converted to electricity globally in 2002 (IEA, 2004). Fossil fuel sources accounted
for 65.3 percent and the rest being mostly nuclear and hydro. Of that total, only 1.9
percent was produced from renewable sources (excluding hydro, including all wind,
geothermal, solar electricity etc.), which comprises 0.31 percent of primary energy
consumption.

Table 1.1 The World’s Proven Fossil Fuel Reserves (IEA 2000)
1995 Consumption Rate of Growth Lifetime (y) Lifetime (y)
Fuel Reserves (*Q)
(Q/y) (%/y) 1987-1997 No Growth with Growth
Coal 24,000 93 0.8 258 140
Oil 9,280 141 1.1 66 50
Gas 6,966 78 2.5 90 50
*Q = 1 Quad = 1 Quadrillion British thermal units (Btu) = 2.9037 E+11 kWh

The figures displayed in Table 1.1 provide an indication of how much longer we can
expect to continue using fossil fuels for energy. The price of oil and gas is likely to
rise dramatically within the next few decades demanding a full scale transformation of
the transportation industry, the highest relative consumer of oil. Once oil and gas
reserves are depleted or become too valuable, the various forms of coal reserves might
serve the population’s energy needs past 2100. If the population is to continue its

24
energy consumption patterns, alternative sources of energy including nuclear fission,
possibly nuclear fusion, geothermal, unconventional fossil fuels and renewables will
become far more important in the future. Realistically from a sustainable perspective,
the relative monetary cost of energy should be far higher than it is now, reflecting its
true cost to humanity. Eventually the relative cost of energy will rise as fossil fuels
deplete. This will place enormous stress on the fundamentals of modern economics,
particularly the necessity for growth.

Australia’s contribution to global energy production is only about 2.5 percent of the
world total (IEA 2004). Per head of population, though, Australia is one of the worst
emitters of greenhouse gasses in the world. The Australian government has failed to
ratify the Kyoto protocol yet it has established a mandatory renewable energy target
(MRET) of generating 9,500GWh by 2010. This is a relatively tiny target and will
mostly be met by existing hydro schemes. According to a report (PMSEIC 2002), the
current government seems highly favorable on continuing to use coal to meet future
energy demands by developing ‘clean coal’ technology. This involves the large scale
gasification of coal and sequestration of CO2 as an effort to suppress climate change,
which is also an outlook of the US government. On a positive note, developing
renewable energy technologies can generate wealth and employment, adding value
within local economies. The amount of funding allocated to develop new renewable
energy technologies with the aim of sustainability, is mostly a political issue, which is
in part a reflection of public opinion in the Australian democracy. Environmental
concerns can change energy policy, which can push growth in renewable energy.

1.5.3 South Australia’s Energy Snapshot

The state of South Australia is one of the most sparsely populated places on earth and
is very rich in renewable energy sources. With 1.1 million people (73% of the States
total), metropolitan Adelaide is the State’s largest energy consumer. SA’s largest
annual generation comes from the power stations at Port Augusta (Northern 530MW
and Playford 120MW) owned by NRG Flinders. They are fuelled by coal mined at
Leigh Creek and transmit power 300km to Adelaide. Burning coal produces the
highest amount of polluting emissions and carbon dioxide per unit energy output
compared to the other major fossil fuels. There are also a few large natural gas-fired
power plants in South Australia including Torrens A&B (1,228MW combined

25
capacity) and Pelican Point (460MW). These gas plants usually operate at only a
fraction of their output capacity, ~37% for Torrens B and ~10% for Torrens A,
whereas the Northern coal plant usually operates at ~93% (ESIPC 2004).

The Electricity Supply Industry Planning Council releases an annual planning report
(ESIPC 2005). It describes in great detail South Australia’s current electricity supply
system, usage, demand and projections so that planning can be made to secure supply.
The following are a list of facts outlining the energy snapshot of SA, many of which
are sourced from this annual planning report.

• Contributes about 6.5% of Australia’s Total Primary Energy Consumption


• Peak electricity demand during 2004-05 is in the order of 3,050MW during
summer around midday, with huge power demand from air-conditioning.
• Average electricity demand is in the order of 1,500MW.
• During 2004 it was calculated about 100 MW or ~7% is lost through heating
transmission cable (ESIPC 2004).
• Olympic Dam mining operations in the State’s north typically consume about
one fifth of the state’s total electricity.
• On average about 15% of SA’s electricity is imported from Victoria, but
sometimes up to 35%, where it is produced cheaply with brown coal (lignite).
• Fossil fuels constituted 98% of all energy used in 2003 (DEH 2005)
• In 1998, less than 0.1% of electrical energy was derived from renewable
sources (DEH 2005); this figure is now approaching 10% of average demand.
• First major wind farm was opened at Starfish Hill near Cape Jervis in October
2003 with a 34.5MW capacity, developed by Tarong Energy and with wind
towers built by Air-Ride in Adelaide
• As of October 2005, with the new Canunda and Wattle Point wind farms there
is 252MW (AusWEA 2005) wind power capacity connected to the grid and
2209MW capacity proposed at different stages.
• SA is currently one of the fastest growing wind farm locations in the world but
is now effectively limited to about 500 MW installed capacity due to a recent
ESCOSA review of wind energy.

26
2 Literature Review and Project Plan

2.1 Renewable Energy Technologies

There are many technologies that can generate useful energy from renewable sources.
Some have been around for hundreds or even thousands of years, but now they are
more sophisticated and built on a larger scale. Other technologies are relatively new
such as photovoltaic (PV) cells. Each technology has its advantages and
disadvantages and they are all at different stages of development. A unique set of
challenges face those who want to generate renewable electricity in a way that it can
compete with conventional fossil fuelled power plants. Currently there are few
renewable energy technologies available that can fairly compete with fossil fuelled
energy. This is because from an economic perspective it is very difficult to put a
monetary value on the intrinsic qualities of things such as clean air or endangered
species and there is currently no cheaper way to produce energy than to extract and
combust fossil fuels, if externalities are ignored.

The aim of a renewable energy project is to make the best use of proven technologies
available in a way that respects the political, economic, consumer market and
technological trends. Ideally once capital, maintenance and repair/replacement costs
are considered, the electricity is essentially provided for free. There are economic
formulas for calculating net present value and IRR, but the answers you get depend on
the gross assumptions that have been made, such as the operating life of installation,
future interest rates, future oil prices and impacts of climate change.

2.1.1 Common Issues with Generating Renewable Power

Unlike fossil or nuclear power, many renewable electricity generation technologies


are unable to run at full output at all times or even for extended periods. Energy from
renewable sources such as direct sunlight, wind and waves is subject to weather
patterns and cycles that cannot be predicted or forecast with great accuracy, only
estimated and averaged over long periods. Wind farms for example have a maximum
design output but rarely produce that value and have zero output in conditions such as
severe or light winds. This is less of a problem for hydro, geothermal and
technologies with energy storage that can provide power on demand.

27
Developers of renewable energy are now finding it desirable to have several
independent sources. Successful connections into the electricity grid are also now
becoming more difficult due to network regulators demanding a ‘high quality’,
uninterrupted, reliable and predictable source. This is why a developer like Wind
prospect wants to explore wave energy generation. When there is no wind there may
be plenty of wave energy, as waves may be generated by winds thousands of
kilometres away (see section 3.2.3 and Figure 3.14).

The term Capacity Factor is often used in the power industry. It is usually expressed
as the ratio of energy produced by a given plant in a year out of the energy that could
be produced by that plant running at full capacity over a year. A low capacity factor
does not necessarily indicate low reliability or lack of efficiency, it simply means than
on average the plant spends little time operating at its rated capacity. When new
renewable energy projects are built the power output capacity is often quoted. It is
important to multiply this value by the capacity factor to get a realistic view of the
plant’s designed average output. For example Starfish Hill has a rated capacity of 34.5
MW but a capacity factor of 34.5% (ESIPC 2004) so it would produce about 12 MW
on average. The “firm capacity” that can be relied on with 95% confidence, has been
estimated at 8% of rated capacity for SA wind farms (ESIPC 2004). According to
OPD, the Pelamis is able to achieve a capacity factor of up to 50%.

Conventionally generated electricity is currently sold to Adelaide residents for about


16-20 cents per kilowatt hour (c/kWh) and the price is expected to steadily rise in
future. Renewably generated electricity can only attract a slightly higher price because
a small but devoted proportion of the population will pay for it. In the Australian
electricity market, “Green Power” can be purchased by the public with guarantees that
it is generated using renewable energy sources. When planning a new power plant, it
is desirable to minimise the payback period of the project by building a plant with the
maximum power delivery for the lowest capital cost. This is where most renewable
energy technologies find it difficult or impossible to compete with new fossil fuel
plants. Plants with long payback periods don’t get built because the financial risk is
just too high and no one is prepared to carry that risk. Often it is a case of whether the
government is prepared to recognise advantages and offer subsidies to the technology.

28
In the case of wind and wave energy, variable supply combined with relatively low
capacity factors has led to restrictions in the amount of power connected to the grid.
Additionally, a grid connection can represent a large proportion of the capital cost of a
wind farm. For a wave farm, the proportion of the capital cost is potentially much
higher still particularly due to the cost of the undersea cable. The submarine cable
between Cape Jervis and Penneshaw is about 13kms long and can carry
approximately 10MW and its value was estimated at about $4-6m in the KI Energy
Review Report (SWWES 2003). ESTA Utilities made an estimation of the total
replacement value being more like $15-20m. This suggests that the cost of a 33kV
10MW submarine cable may cost up to $1m per kilometer, compared to the figure of
a 33kV land cable being $50,000-$80,000 per kilometer (SWWES 2003). It is clear
that a wave farm should be as close as practicable to the shore, consistent with a good
wave resource.

Some renewable energy technologies require rare materials or large amounts of


materials and even require energy intensive manufacturing. If many of the current
renewable energy schemes were used to provide power for hundreds of millions of
people, vast expanses of land and gigantic amounts of resources would be needed.
Biomass has a limit to the amount of power that can be generated sustainably for the
area of land used (~0.5W/m2 thermal, see Table 2.1) and environmental damage must
be closely monitored. For other renewables, land usage this isn’t such an issue: for
example wind turbines don’t prevent cows grazing under them. Photovoltaic solar
panels require the use of silicon crystals that are expensive to grow and cut, which has
limited their wholesale use. The Pelamis does not occupy land as such and but will
require several hundred tons of steel per MW. This project includes an investigation
into all these issues specific to the Pelamis in the Life Cycle Analysis in Chapter 6.

Some types of renewable energy projects have either been protested heavily or even
stopped because of their adverse environmental impacts, direct or indirect.
Construction of hydroelectric dams, such as those in Tasmania, left environmentalists
enraged by the destruction of vast areas of wilderness and this was the centre for
ongoing political turmoil. Aside from environmental issues, aesthetics can also be a
source of disturbance. Wind farms have been criticised for visual pollution, noise

29
pollution and effect on birdlife (and bats). The Pelamis has a very minimal direct
environmental impact; see section 6.3 and APPENDIX K.

2.1.2 Overview of Renewable Energy Technologies

This section contains an overview of all established or promising renewable energy


technologies for use in generating electricity on a relatively large scale. These are the
technologies that have been successful enough to evolve to the stage of being
deployed in big engineering projects. Key factors including the scale, effectiveness
and state of technological advancement are outlined for each technology. This is done
in order to make simple direct comparisons between each technology and wave
energy technology, in particular the Pelamis. Given that the Pelamis is a very recently
deployed technology, it is important to asses where wave energy stands within the
renewable energy industry.

Biomass is the most heavily consumed form of renewable energy and is listed in
APPENDIX A as combustible renewables and waste. It includes all forms of recently
produced matter as a result of photosynthesis. It wasn’t long ago when biomass was
mankind’s primary energy source. Most of this energy source is used in combustion
processes to produce heat, generating CO2 and other emissions. A very small
proportion of total biomass energy is currently being converted to electricity. Biomass
is completely renewable but if it were to replace all current fossil energy use, it would
be far from sustainable using current proven technologies. Biomass can be converted
into other fuels such as biodiesel or alcohols on a large scale, many of which offer
cleaner emissions than petrochemicals. Biomass energy extracted from waste will
become more important in the future.

Hydro power is the only renewable source of energy that produces electricity on a
large scale today. It represented 89.5% of all renewably generated electricity
produced in 2002 (IEA 2004). Hydro electricity is sourced from the natural
evaporation of water by the sun. Principally the sun performs all the work in lifting
millions of tonnes of water above sea level and depositing it in hydroelectric dams.
The water is then released through water turbines, which turn generators. Dams are
built across flowing bodies of water in order to raise the hydraulic head and act as
energy storage. Hydro’s main advantage is producing power quickly at any time of

30
demand with minimal emissions. It is now becoming clear that some dams produce
considerable greenhouse gas emissions, methane in particular (Boyle 2004).
Geological and terrain features limit the amount the hydropower can be exploited and
most of the world’s largest hydropower sites are, or are soon to be, in use. A giant
hydropower plant is under construction in China and is due to be finished in 2009.
The Three Gorges Dam on the Yangtze River will be a mile wide, 172m high and able
to generate 18,200MWe (IRN 2005), making it by far the biggest single renewable
electricity project in the world. Its construction will require the demolition of several
towns and the displacement of over 1.9 million people (IRN 2005). Many believe the
dam will help prevent chaotic flooding of the Yangtze, such as the disaster in 1998
resulting in the evacuation of millions and a staggering death toll. Others think that it
may only compound the problem; however this issue was certainly a major factor that
went into approving its construction.

Extracting geothermal energy is essentially the process of mining geothermal heat. It


is the only energy source that is classified as renewable that supplies constant non-
cyclical energy independent of the sun. Global geothermal electricity generation
capacity approached 8 GWe in 2000 (Boyle 2004) and a further 16GWt heat was
extracted for a variety of processes. Deep within the earth there is a vast amount of
energy, almost limitless, but on the surface there are definite limits to what can be
feasibly extracted. The best sites for ‘high-enthalpy’ geothermal power generation are
typically located at boundaries of tectonic plates or fault lines. Aside from volcanic
activity, ‘hot dry rocks’ created partly by radioactive decay are now being explored in
detail. Central Australia has such a resource in great abundance. The technology
deployed is simple in concept, one or more boreholes are drilled into the heat
‘reservoir’ and then heated fluid is pumped to the surface to conventional steam
turbines or heat exchangers. However it is located very far from urban centres of
consumption and the deep drilling required is very expensive. It could however be
used as an energy source for regional Australia and in energy intensive industrial sites
such as Olympic Dam, SA.

31
Wind power has been used for thousands of
years for sailing ships and windmills. Winds are
driven by the sun’s differential heating of the
ground. Modern methods of large scale
electricity generation from wind utililise a
number of efficient wind turbines combined in a
‘wind farm’. The turbines are massive structures;
manufacturers are now producing turbines with
hub heights and rotor diameters over 100m. One
such machine, the V120 developed by Vestas
Wind Systems, has a rated output capacity of
4,500kWe, see Figure 2.1. Larger wind turbines
can collect more energy per disk area due to
greater wind velocities available higher from the
surface. Wind power is one of the most rapidly
Figure 2.1 (Vestas 2004)
advancing energy markets and is experiencing a
growth in development projects worldwide. In the 70’s and 80’s there were a range of
different wind turbine designs and configurations on the market, including vertical
axis wind turbines. Eventually the 3 blade horizontal axis turbine design matured and
established itself as the market leader. Recently there has been a lot of investment in
offshore wind energy particularly in Europe. As of writing this report the world’s
largest approved wind farm will be constructed off the shores of Ireland on the
Arklow Sandbank. With 200 turbines, it will have an output capacity of 520MWe.

All life on earth depends on energy received from the sun. Our planet continuously
receives solar energy at about 9,000 times the current total primary energy
consumption (IEA 2002). Of course this amount is far greater than can be technically
be captured and harnessed. In the tropics up to about 1kW/m2 makes it to the surface
at noon. Solar electricity represented only about 0.05 percent of total electricity from
renewable energy sources in 2000 (IEA 2002). Two groups of technologies,
photovoltaic (PV) ‘solar panel’ systems and solar-thermal power plants, each have a
roughly even share of that production. Energy produced by PV systems is still
relatively expensive but has advantages in small plants in remote areas. Large solar
power plants use solar-thermal technologies that generate power from the heat of solar

32
radiation. As with most coal, nuclear or geothermal power plants utilizing heat
energy, electricity from concentrating solar thermal makes use of the Rankine cycle
(or variations) to provide work for steam turbines. This process has inherent losses
and limits of efficiency in the order of 40%. The concept behind the three main solar
thermal technologies is illustrated in figure 2.2.

Figure 2.2 Concentrating solar thermal collector types (Johansson et al. 1993)

Each of these configurations makes use of reflective mirror-like surfaces to


concentrate the suns energy. The largest solar power plant currently in operation is the
SEGS built by LUZ international situated in the Mojave Desert, California. The SEGS
is a 100 acre (0.4km2) field of parabolic trough mirrors and can generate a maximum
power output of 355MWe. This plant currently constitutes nearly 90% of the world’s
direct solar electricity; however its capacity factor is about 22% (Hayden 2001).
There is research being undertaken around the world using the parabolic dish
configuration to power advanced Stirling engines and high temperature photovoltaics.
In this way some of the highest levels of direct solar capture efficiency have been
achieved, over 30% (peak of about 300W/m2 of collector) at temperatures as high as
1000°C (Boyle 2004).

The ocean has very large sources of renewable energy that have remained virtually
unexploited. This is because the energy is often widely dispersed and located far from
where consumption takes place. Tidal or ‘lunar’ energy moves great bodies of water
through many different cycles and is produced by the gravitational attraction of the
moon and sun. The most well established technology uses a tidal barrage that is like a
big, low head, hydroelectric dam which use turbines to generate significant amounts
of electricity. The largest tidal barrage was built in France in 1966 and can generate
up to 240MWe. Tidal barrage projects can be very costly and can and adversely effect

33
the environment. Ocean tidal current turbines are like under water wind farms and
would counter these problems. The energy density is often too low to become cost
effective, but tidal turbine technology is advancing. Ocean thermal energy conversion
exploits the temperature differences between the surface and the floor of the sea. This
is a giant source of solar thermal energy but due to such poor energy density,
enormous structures would need to be built to produce sizeable amounts of power.
Other interesting ocean energy technologies include exploiting salinity gradients by
osmosis of fresh water meeting the sea; again this has low energy density.

Table 2.1 Annually Averaged Extractable Energy Flux (Fay & Golomb 2002)
Source Area Heat (W/m2) Work (W/m2)

Solar thermal Collector 150 20


Solar photovoltaic Cell 30
Wind Turbine disk 40
Tidal current turbine Frontal area 750*
Ocean wave Frontal area 10,000+ (W/m crest)
Biomass Field 0.5 0.1
Geothermal Field (underground) 0.1 0.02
Tidal barrage Tidal pond 1
Hydropower Drainage basin 0.01
* This is an estimation for Backstairs Passage, South Australia (Kirke, pers. comm)

Each renewable energy source has a feasibly extractable energy flux or power
available from a certain area of resource. A comparison of these values averaged
annually is provided in Table 2.1. The first five sources in this table can be compared
relatively easily regarding energy density and give some indication of the area of
structure required for collection. The last four sources in this table are a little more
difficult to compare and greatly depend on location, geography and a number of other
factors. For these four sources typical values have been provided for the surface area
of the resource. Low energy flux values help explain why hydroelectric dams & tidal
barrages typically need to be such giant structures to produce large amounts of power,
even if they don’t cover the whole basin. Solar and wind power have comparable
energy density although wind power is currently significantly more cost effective and
has the advantage of extracting energy already in kinetic form.

34
Waves alone carry enormous amounts of power but their exploitation is in its infancy.
What makes wave energy stand out from the rest is the inherently high energy density
available. This high energy density suggests that a large amount of energy could be
extracted for minimal resources given the right technology, although in practice it is
not so simple. The wave energy resource and technologies will be covered in detail
throughout the following chapters of this literature review.

2.1.2 Energy Storage and Transportation Technology

Technological developments for the storage and transportation of energy could have a
great effect on the energy industry and society as a whole. Energy in the form of
electricity, arguably the most useful form of energy, is very expensive to store and
transport efficiently. A revolution in energy storage or transportation could have
numerous dramatic effects on the future of energy use, on all forms of technology and
could greatly expand the renewable energy industry.

In the case of South Australia and in many other places in the world, we are quickly
approaching the limit of localised total power capacity from wind farms that can be
connected to the grid due to the variability of supply. Large insertions of ocean wave
generated electricity would also be restricted for the same reasons as wind energy. If
all of the power generated from wind or wave farms was able to be directed to large
scale energy storage so that all the power is available on demand, then this problem
would be largely solved. In fact a huge amount of energy is wasted all over the world
from large scale power plants and industrial use, right down to transport and
household use due to inadequate energy storage. Big fossil fuel power plants (>1GW
capacity) must be fired up in anticipation of load and a large amount of energy is lost
before the load arrives. Gas turbines can come online in 30 to 60 minutes, while steam
turbines are much slower (Kirke, pers. comm). Improvements in energy storage and
transportation could greatly extend the life of our fossil fuel reserves, remove the
threat of power outages and allow much more effective use of renewable energy –
securing our energy future.

Energy can be stored in many ways using all kinds of physical and chemical
properties of matter. New research and technologies have led to advances in many

35
forms of energy storage including thermal storage, chemical batteries (Vanadium
Redox battery technology is promising for large scale), ultra-capacitors, salinity
gradients and compressed gas systems. Two other promising forms of energy storage
technology involve using advanced flywheels and hydrogen fuel cell technology and
will be discussed briefly here. In concept neither is new, but both technologies are
rapidly advancing and can provide compact and high quality energy.

A spinning flywheel is a compact form of kinetic or mechanical energy storage. For


high density energy storage the flywheel’s material tensile strength relative to its
density is most important, not so much the weight or the size of the flywheel. The
most advanced flywheels make use of modern composite materials shaped with high
precision and rotate at extremely high velocities. To minimise losses, the flywheel is
suspended by a magnetic bearing and enclosed in a chamber with a partial vacuum or
an inert low viscosity gas. Coupled to the shaft are the latest high power and
efficiency brushless motors and generators. The result is a unit that offers many
advantages over batteries including being able to deliver a wide range of power, with
relatively tiny losses for an almost indefinite number of cycles. NASA is even
researching their use in spacecraft (NASA 2005). The cost of this technology is still
very high for the amount of energy stored, mainly due to the flywheel material and the
high powered magnets. If the cost approached that of conventional batteries, units
could be purchased for individual household or industrial use. At this scale they could
make much better use of renewable energy by ‘smoothing out’ and improving the
quality of supply from cyclical energy sources such as direct solar. Batteries for small
scale photovoltaic power plants are a continually expanding market.

Much speculation exists about the prospects of a future ‘hydrogen economy’ where
pure hydrogen energy is to become a major component of primary energy,
particularly for transportation. Extensive research is being undertaken into producing
hydrogen gas and then converting it to electricity using different types of fuel cells or
even combustion in turbines and internal combustion engines. It must be stressed that
Hydrogen is not a source of fuel as it does not occur naturally in a pure state and
always requires more energy to produce than can be extracted from it. The attraction
lies in the fact that fuel cells can convert hydrogen into energy in a way that is very

36
efficient, clean and environmentally friendly and that hydrogen is the most abundant
element in the universe.

Figure 2.3 Current Hydrogen consumption (Kruse, Grinna & Buch 2002)

Global production of Hydrogen is over 45 million tones per year and over 90% of this
is from reformed raw fossil fuels (Kruse, Grinna & Buch 2002). The distribution of
Hydrogen consumption is shown in Figure 2.3 where the majority of it goes to
Ammonia production, mainly for the use of fertilizer, and most of the rest goes
towards refining fossil fuels. Ideally Hydrogen could be produced on a large scale
from the electrolysis of water using renewable electricity. The leading technology for
this is alkaline electrolysis where manufacturers are now quoting efficiencies of 4.2
kWh/Nm3 (Stuartenergy 2005), which equates to an energy return on energy invested
(EROEI) of 71%. There is considerable research taking place on using thermal energy
to dissociate water to obtain hydrogen, also photoelectrolysis and biological
production. Proponents of nuclear power expect that the hydrogen economy will be
largely nuclear driven and thermal dissociation is a potential means of production.

Even though hydrogen is being discussed here as storage for renewable energy, the
storage and transportation of hydrogen is regarded as a major flaw of the hydrogen
economy. Although it has very high energy density, its volumetric energy density is
very low, see Table 2.2. Hydrogen must be compressed to achieve acceptably high
energy storage density. Liquid hydrogen requires a cryogenic temperature of -253°C
needing up to 40% of its energy content (Kruse, Grinna & Buch 2002). Large scale
storage could be achieved underground under pressure. Natural gas infrastructure
could be upgraded to store and deliver hydrogen at significant cost.

37
Table 2.2 Energy Densities of Various Fuels (from various sources)
Fuel State Thermal Energy Volumetric Energy
Density (MJ/kg) Density (kWh/L)
Hydrogen -253°C (liquid) 120 2.33
Ammonia -33.5°C (liquid) 20.9 4
Gasoline (liquid) 1 bar, 20°C 43.2 8.6

Issues with hydrogen storage for transportation have limited the growth of the
industry. Many compact storage solutions have been proposed for vehicles, including
compressed gas tanks, cryogenic LH2 tanks and storage in solids. Studies have been
done by Chahine & Bernard (2001) and Pradhan et al. (2002) suggesting there may be
storage solutions in using carbon nanofibre technology. The cheapest and most
effective storage of hydrogen is with its natural energy carriers, the hydrocarbons
(especially methane), alcohols and ammonia.

The idea of using ammonia, one of the most mass produced chemicals in the world, as
a “carbon free” hydrogen carrier is not new. A great deal of infrastructure is already in
place to store and distribute ammonia, which in a liquid state (needing only -33.5°C
or 8 bar) has half the volumetric energy density of gasoline (Kordesch et al. 2003). A
simple ammonia cracker and an alkaline fuel cell can be used to provide electrical
energy or it can be used directly in specially designed internal combustion engines.
The main drawbacks are that it is toxic and that the energy needed to produce
ammonia from hydrogen by the Haber Process is approximately 70% of its energy
content.

In the short term, mass production of Hydrogen could occur through the gasification
and reforming of fossil fuels, especially coal or even biomass, and then controlling
CO2 emissions by sequestration technology. From there efforts could be made to
develop the required infrastructure for hydrogen manufacturing and supply using
renewable or possibly nuclear energy.

Hydroelectric dams are currently unparalleled in their ability to store large amounts of
renewable energy with minimal losses and can respond to demand in less than one
minute. The idea of transporting excess energy into pumped hydro storage has great
merits if excess capacity is available. Building renewable energy plants with

38
fluctuating output such as wind turbines near dams could improve their economic
attractiveness by evening out supply and therefore increasing their effective capacity
factor. In SA pumped storage is the only economically feasible form of large scale
energy storage. Using Kangaroo Creek Dam and an enlarged dam at Paracombe,
about 50MW could be generated for 6 hours (Kirke, pers. comm). Also a new dam
built below the existing dam at Myponga could supply up to 600MWh with a 50-
100MW capacity at a cost of approximately $1.2m/MW. Losses involved in
recovering the energy used to pump to a dam would be in the order of 25-30%. SA
has few dams, but elsewhere in Australia and the rest of the hydro electric world there
would certainly be other worthy and beneficial opportunities for pumped storage.

39
2.2 The Power of Sea Waves

The waves in our oceans are very effective at storing, transporting and concentrating
large amounts of energy. A strong yet brutal example of this ability is the devastation
caused by the Asian tsunami disaster of December 2004 created by powerful
earthquakes beneath the Indian Ocean hundreds of kilometers from shore. With the
exception of geologically driven tsunamis, almost all of the energy in ocean surface
waves originated from the sun. Anyone who has experienced conditions in the open
ocean, or witnessed coastlines during severe weather holds an appreciation of the
power of ocean waves. Estimations have been made for the feasibly extractable global
wave power resource from 1TW (Johansson et al. 1993) to 2TW as estimated by the
World Energy Council, mentioned in Duckers (2004). Considering the current global
electricity demand has just exceeded 1TW (IEA 2004), the power available in waves
is a very large untapped resource. Given enough technological development these
figures could be attainable.

This section contains a review of wave energy resource theory; an in-depth study of
southern Australia’s wave energy resource is undertaken in Chapter 3. APPENDIX C
contains a compilation of all wave energy information resources used throughout this
project (excluding energy converter technologies).

2.2.1 Oceanic Wave Formation and Climatology

Waves on the surface of the ocean are generated by sea winds that were generated by
solar energy. Wind energy represents less than one percent of the solar energy that
arrives at our planet and only a tiny fraction of that energy is transferred to waves
(Boyle 2004). Exactly how this process is initiated is complex and not completely
understood. There are three processes that describe the formation of ocean waves
(Duckers 2004).

1. Air flow creates tangential stresses on the surface of the water,


spawning the initial growth.
2. Air flow becomes turbulent at the water surface and wave development
is caused by fluctuating stress and pressure oscillations in phase with
existing waves. These are called capillary waves.

40
3. Growth continues on waves after they have reached a certain size from
wind exerting strong forces on the upwind face of the waves.

Waves that have traveled out of stormy waters are known as ‘swell’ waves. Storms
generate waves that can travel great distances with minimal loss of energy in deep
oceans. Three factors contribute to the size of waves produced: the wind speed, its
duration and the distance over which the energy is transferred to the waves, called the
‘fetch’ (Duckers 2004).

Figure 2.4 Illustration of wave fetch area to Cape Sorell (Reid & Fandry 2004)

The majority of waves arriving at South Australia’s coasts would capture most of
their energy over a fetch area not too dissimilar to that shown in Figure 2.4 for Cape
Sorrel on the west coast of Tasmania. A study of waves arriving at Cape Sorell is
provided in section 3.3. In theory the fetch distances for waves arriving at many
locations in the world would stretch several thousands kilometres, sometimes more

41
than half way around the Earth. The difference between short fetch ‘chop’ created by
local winds and long range swell can be identified in the images in APPENDIX D.

It is possible to estimate the long term wave climate of any part of any coast given
enough data and analysis. The meteorological study of wave climate is a complex
process involving many factors. Wave energy is strongly variable in time and space,
involving a lot of chaos; energy levels can be estimated and predicted as accurately as
weather can be. An important factor to determine is the directional climate of long
fetch swell that contains most of the energy. At present there are a number of
computer models available which supply calculated information about waves at any
time and location in the ocean. Some of the most well established models are the
NOAA Wavewatch III and the ERA-40. Details of these and other models are listed
in APPENDIX C. Data input for these models include a host of information collected
from satellites, also wind data (from anemometers etc.) and wave rider buoy data
collected from various sites around the world. Calculations are made by the models
which make correlations between each of the data inputs and the measured wave rider
buoy data.

Figure 2.5 Annual average wave power in kW/m of crest length, for various
locations around the globe (Duckers, 2004).

42
The parts of the world which have the highest wave energy resources are also
subjected to regular high wind velocities. Some parts of the globe have wave energy
levels that regularly exceed 100kW/m. In Figure 2.5 it is apparent that there are higher
wave energy levels available closer to the poles and on the western coasts of
continents. This phenomenon is due to the powerful airstreams that constantly circle
the poles west to east created in part by the direction of rotation of the Earth on its
axis and the flow of solar energy over the surface. Figure 2.6 is a representative
sample showing the typical clockwise wind direction and velocities (m/s) when
looking over the southern pole.

Figure 2.6 Monthly mean 500 hPa wind speed (m/s) and direction, Southern
Hemisphere, July 2005.
(Source: Australian Bureau of Meteorology, www.bom.gov.au 2005)

43
2.2.2 Wave Energy Characteristics

Ocean waves are complex phenomena


and rarely resemble perfectly
monochromatic or smooth sinusoidal
shaped waves. They are usually
described as gravity waves or waves
that are mainly controlled by gravity
and inertia. At higher amplitudes for a
given wavelength the peaks of ocean
waves tend to become narrower as
illustrated in Figure 2.7 (Nave 2005).
Experiments done in wave tanks
(Soreneson 2000) show that the limit
of amplitude against wavelength, or

Figure 2.7 Gravity Waves (Nave 2005) steepness, is about 1 to 7 and a


minimum wave peak angle of 120°.
Beyond this the top of the wave is traveling too fast and becomes unstable then
‘breaks’. Waves observed in the oceans are typically composed of many gravity
waves traveling in various directions. Much of the surface of the waves is covered in
newly forming capillary waves from wind turbulence. Large and powerful long swell
waves will travel hundreds kilometers to the shore through short fetch waves traveling
in any direction. This is best shown in Image A of APPENDIX D.

The physical properties of sinusoidal waves can be used to model ocean wave
characteristics. Most of the theory outlined here is sourced from literature by Duckers
(2004) and Fay & Golomb (2002). Wave power is normally expressed as the amount
of power per length of wave crest (kW/m). Figure 2.8 illustrates an idealized
sinusoidal ocean wave and its characteristics.

44
H = wave height
λ = wave length
d = mean water depth
T = period
ρ = density of water (1000kg/m3)
g = gravitational acceleration (9.8m/s2)

Figure 2.8 An idealized sinusoidal ocean wave.

ρg 2 H 2T −1
Wave power/length = P(Wm ) = (2.1)
32π
The above equation gives the wave power for an idealized ocean wave. It shows that
the power of a wave is proportional to the square of its amplitude. It is also useful to
know that 95% of the energy of a wave is contained between the surface and a depth
equal to a quarter of the wavelength (d = λ/4).

Oceanographers have developed the following equations that relate wave properties
empirically to wind speed, V (Fay & Golomb 2002).
2 V4
H = 2.19 E (−2) 2 (2.2)
g
2π 0.70 g
= (2.3)
T V
These two relations can be substituted into equation (2.1) to give:
ρV 5
Wave power/length = P(kWm −1 ) = 3.89 E (−3) (2.4)
g

200

175

150

125
P (kW/m)

100

75

50

25

0
4 5 6 7 8 9 10 11 12 13 14
V (m/s)

Figure 2.9 Plot for equation (2.4) empirically relating wind speed to wave power

45
In Figure 2.9 a quick plot generated using Excel is displayed for equation (2.4); it
shows a strong exponential trend. This was done to visualize how this empirical
relationship was derived. This type of formula would have great accuracy limitations
in its application because fetch distances vary and wind rarely flows in a straight line
for long. It would have its uses in determining simple correlations when initially only
wind data is supplied, such as wind strengths for an upcoming storm.

The greatest wave energy levels exist in deep water where waves travel at speed with
minimal losses. Once they move into depths of about 50m the energy losses from
bottom friction begin to accumulate resulting in a reduction in wave velocity. Figure
2.10 shows how particle motion below the surface can interact with the seabed in
shallow water. In deep water the particle orbits are circular, but in shallower water the
orbits are elliptical applying frictional forces to the seabed. All kinds of factors and
wave properties come into play as waves interact with shoreline bathymetry, making
for complicated modeling. Eventually all of the energy is lost on the shore when the
waves break and then interact with themselves. As wave speed reduces the effect of
energy conservation increases their amplitude, bringing about the phenomenon
described in Figure 2.7. It is for these reasons that offshore wave energy (>50m
depth) devices have a clear advantage over shoreline device. Offshore wave energy
converters can access more energy, more predictably and without being subjected to
the rough forces of breaking waves.

Figure 2.10 Illustration of approximate particle orbits for: a) deep water where
λ ≈ H, b) in shallower water where λ ≈ 2.5H (Brooke 2003)

46
Another important phenomenon of waves moving into a coastline is known as
refraction. Interactions with coastline bathymetry give waves a tendency to arrive at
the shore with the crest parallel or close to parallel to it. This is created by reduced
wave velocities forcing a change in directional orientation over a length of crest.
Images B and C of APPENDIX D both illustrate this property. Depth contours affect
the wave velocity of long fetch waves and concentrate their energy at headlands,
capes or points. These headlands also have a great sheltering effect for regions in the
shadow of the long swell direction. Such areas are fetch limited and have much
reduced wave energy levels that are difficult to model. The importance of this will be
seen in the KI Case Study in Chapter 4.

Two important values that will be used throughout most of Chapter 3 are those of
significant wave height (Hs) and zero crossing period (Tz). The surface of the ocean is
irregular and composed of a number of different waves that can’t be measured
individually. In order to measure them an averaging process is required. The average
water height will always be zero and this is used as reference point for other
measurements. Squaring all instantaneous elevations makes them positive, and then a
mean height of a given number of waves is taken. The significant wave height is
calculated as the square root of this mean multiplied by 4 (Duckers 2004). Through
observation, the value for significant wave height closely approximates the average of
the highest one third of the waves. The zero crossing period is given by the average
time between upward movements of the wave surface thought the mean level. An
example of how these values are calculated is illustrated in Figure 2.11.

Figure 2.11 A typical record from a Wave Rider Buoy (Duckers 2004)

47
The significant wave height for this example is Hs = 3m and the number of upward
crossings in 150 seconds is 15, giving Tz = 10 seconds (Duckers 2004). Once Hs and
Tz are determined, equation (2.5) can be used to determine the average total power per
length of crest.
H s2Tz
Average Power/length = P(kWm −1 ) = (2.5)
2
This is one of the most important equations used in this project and is used
extensively throughout Chapters 3 and 4 to determine available wave power. For the
example above the average wave power can be determined for the period the data was
measured:
H s2Tz 32 × 10
= = 45kWm −1 (2.6)
2 2

Again it can be seen that the power of a wave is related to the square of its height. The
next section deals with how to harness this dense but highly variable energy source.

48
2.3 Wave Energy Converters

Wave energy was first explored in detail as a major source of energy during the
1970’s energy crisis. Most of the investment into wave energy research has been in
Europe and this is where many of the wave energy converters have been developed.
This is because Europe has become increasingly reliant on importing its energy and is
now focusing on renewables in an effort to improve energy security. Wave energy
converters are currently still going through an evolutionary stage in much the same
way wind turbine technology did in the 70’s and 80’s. A different set of challenges
apply to wave energy converters and many of the technologies are still evolving fast.
Figure 2.12 shows the evolution of some wave energy technologies on the basis of
cost per performance output. A number of different contending technologies are
named, with the Pelamis currently taking the lead. These figures are estimates by
Duckers (2004) from a range of sources, but they provide a clear trend that wave
energy devices are dropping in cost. Duckers (2004) provides a concise overview of
different wave energy converters.

Figure 2.12 Evolution of UK wave power devices (Duckers 2004)

2.3.1 Study of Different Designs

Every wave energy converter consists of some form of structure that has the goal of
extracting the mechanical energy from the waves. There are devices built into the
shoreline or offshore devices which are typically designed to be located in depths
greater than 50m where losses of long swell energy are minimal. Offshore structures
can be rigidly or flexibly moored to the sea bed or the entire vessel can be designed to
float. In order to extract a wave’s energy there must be some relative movement

49
between the structure and the wave motion. A number of mechanisms have been
devised for this purpose. A diagrammatic illustration of the main families of
configurations is shown in Figure 2.13. Wave energy converters can be classified into
5 different groups, but for some designs this can be difficult task.

Figure 2.13 Configurations of Wave Energy Converters (Johansson et al. 1993)

Tapered channels can be effectively built into the shoreline and work by concentrating
waves up a channel to create a static hydraulic head in a storage reservoir. Much like
a small dam, the water then returns to the sea through a turbine built into the structure.
A surge device using the flexible bag and spine mechanism was developed by
Coventry University in the 1980’s called the circular Clam.

50
Many different technologies have been devised using the oscillating water column
(OWC) principle. One study (Irvin, Thiagarajan & Thomas 2005) undertaken in a
wave tank suggests the hydrodynamic efficiency of a fixed platform OWC is in the
order of η=70%. The OWC is usually sealed and power take off is achieved by a
pneumatic turbine. An ingenious design has been employed to take advantage of the
bi-directional flow created by OWCs called the Wells turbine. It is self-rectifying,
meaning it can accept air flow in either axial direction while rotating in one direction.
Part load efficiency is a problem of most wave energy converters that are designed to
operate at optimum parameters. Improvements in part load efficiency of self
rectifying turbines have been achieved by utilizing flexible or variable pitch blades
(Beyene & Wilson 2005).

The Edinburgh Duck developed by Professor Stephen Salter, also known as the Salter
Cam, is one of the most theoretically efficient wave energy converters available
(Duckers 2004). It is classified as a pitching device in Figure 2.13 and could be
envisaged as a number of cams linked together by a long flexible spine. A large scale
device could be tuned to match the orbital motion of long swell waves, leaving almost
calm sea in its wake. This concept has not been developed on a large scale because
energy extraction presents a number of engineering challenges. The Pelamis has its
ancestry in the Edinburgh Duck (Duckers 2004). It is a pitching and heaving float (see
Figure 2.13) where all of the work is accumulated within the structure at the joints.
Further description of the Pelamis design is included in section 2.4.1.

Another way to classify wave energy converters is by their geometry and orientation.
• Attenuators have their principal axis perpendicular to the wave front, such as
the Pelamis.
• Terminator devices have their principal axis parallel to the incident wave
front; an example is the Edinburgh Duck.
• Point absorbers are the other form, exemplified by tethered buoy systems (see
heaving float, Figure 2.13).

Many criteria could be used to evaluate all of these wave energy converter
technologies. In practice one of the most important design aspects is survivability.

51
Ocean waves repeatedly apply rough and highly variable loads to wave energy
converters and some expensive prototypes have been destroyed as a result of powerful
storms. In the end the most important thing to determine is the commercial viability
and readiness for a given technology. A thorough assessment of all available offshore
wave energy conversion devices has been undertaken in the US by the Electric Power
Research Institute, or EPRI (Prevesic 2004). The report, released on June 16 2004,
identified only 8 technologies on the market ready for demonstration by 2006 and
assessed them using an extensive list of criteria; design for survivability being a key
requirement. In summary the EPRI believed only one of the devices was acceptable
for use in a pilot plant in the US, the Pelamis by OPD. Three other devices, namely
Energetech (see Figure 2.14), Wave Dragon and Waveswing could be used if a few
remaining issues were addressed.

2.3.2 Projects at Advanced Development Stage

On May 19th 2005, OPD made the announcement of the world’s first commercial
wave farm to be located 5km off the Portuguese coast. Three Pelamis are to be
installed having a capacity of 2.25MW and is expected to meet the average electricity
demand of more than 1,500 Portuguese households. “Subject to the satisfactory
performance of the first stage, an order for a further 30 Pelamis machines (20MW) is
anticipated (OPD 2005)”. The cost of the first stage of this project is €8m
(~$AUD13m). This is good indication that the cost of a locally developed Pelamis
would be in the order of $AUD4m per unit until more units are ordered for production
significantly reducing the cost.

Ocean Prospect in the UK is working on developing a wave farm for a wave energy
project called the Wave Hub. “The Wave Hub concept is to build an electrical grid
connection point approximately 10 miles off the UK shores into which wave energy
devices would be connected. It will provide a well defined and monitored site with
electrical connection to the onshore electricity grid and will greatly simplify and
shorten the legal consents process for developers. Wave Hub would reduce the risk
for developers of the first pre-commercial wave machine arrays. If it goes ahead,
Wave Hub could be commissioned by 2006 (OPD 2005)”.

52
The leading Australian-designed wave energy converter has been developed by
Energetech. A demonstration project known as the wave energy barge (WEB), see
Figure 2.14, has been completed on the breakwater at Port Kembla on the coast of
NSW with a projected output capacity of 500kW (Energetech 2005). A $750,000
grant was secured to develop this prototype which is currently undergoing testing. On
the 26th of October the capacity was successfully demonstrated and exceeded
expectation. It is the first time a floating OWC concept has been demonstrated and
using the new Dennis Auld rotating turbine blade technology. The steel structure is
35m wide by 36m long and weighs a total of 485 tons. Energetech have also recently
announced that they have incorporated reverse osmosis desalination units with their
Port Kembla project. They have developed a partnership with Wollongong company
H2AU and claim their project could deliver 3 million litres of fresh water per day (3
MLD). This figure seems too large from the research undertaken in Chapter 5.

Figure 2.14 Energetech’s WEB Design (Energetech 2005)

53
2.4 The Pelamis Wave Energy Converter

After 6 years of extensive research and development, the Pelamis P-750 Wave energy
converter developed by Ocean Power Delivery is ready for commercialization.
Looking rather like a string of 4 gigantic hotdog sausages, in design concept it is
unique from any other developed wave energy converter, indeed any other power
plant. OPD have focused all their efforts and business on this one technology and
claim at least an 18 month lead over rival companies. The technology has been tested,
proven and verified by WS Atkins, an independent engineering consultancy. This
section provides only a brief description of the technology as a there is copy of the
Pelamis brochure in APPENDIX A containing a concise overview. Further
investigation into the Pelamis design is available in Chapters 5 and 6.

2.4.1 Design Concept and Features

The word Pelamis is Latin for serpent; the Yellow-bellied Sea Snake has the species
name Pelamis platurus. In concept the Pelamis resembles a giant sea snake floating
on the surface of the ocean. It uses hydrostatic forces of gravity versus buoyancy to
collect energy within its own structure without a rigid support, such as the sea floor or
cliff face. This is a key design features as all of the other wave energy converters on
the market need a rigid support. It only needs flexible mooring at depths from 50 to
100 meters. The structure ‘weathervanes’ so it always points into on-coming waves.
Survivability is a key design theme for the Pelamis as OPD recognizes that the most
efficient or effective wave energy converter in the world is useless if it can’t
withstand stormy conditions. There are no experimental manufacturing techniques or
components used in making a Pelamis, as it makes use of 100% proven, available
technology from the offshore oil and gas sector. The hull of a Pelamis can be
fabricated by manufacturers of wind turbine towers. Final assembly can be undertaken
while the sections are submerged. The Pelamis can be rapidly attached or detached for
towing using the yoke at the nose. During operation the yoke hangs down and acts as
a stabilizer, correctly orienting the Pelamis and preventing it from rolling over.

54
2.4.2 Performance and Specification

A list of important specifications for the Pelamis is located on the last page of the
brochure in APPENDIX A. A power matrix is supplied in the brochure for the power
period (Tp) derived using an experimentally verified numerical model. For use in this
project another slightly more detailed performance matrix was supplied, Figure 2.15,
for different values of significant wave height (Hs) and the zero crossing period (Tz).
The highlighted values show output figures that exceed the maximum output of
750kW. In practice the absorbed power would be limited to 750kW but these values
are an indication of the theoretical unrestricted output. This matrix will be used in
Chapter 4 to model the Pelamis performance at Cape Du Couedic, Kangaroo Island
and Cape Sorell, Tasmania.

Wave Period (tz, s)


0.0 3.5 4.5 5.5 6.5 7.5 8.5 9.5 10.5 11.5 12.5 13.5 14.5 16.5 19.0
0.25 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
0.75 0.00 0.00 28.77 33.14 28.58 22.04 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
1.25 0.00 32.84 79.91 92.07 79.38 61.21 45.47 33.61 25.05 0.00 0.00 0.00 0.00 0.00
1.75 0.00 64.37 156.63 180.45 155.58 119.97 89.13 65.88 49.09 37.06 28.36 22.01 0.00 0.00
2.25 0.00 106.41 258.93 285.06 246.96 198.32 147.34 108.91 81.15 61.26 46.89 36.38 22.68 0.00
2.75 0.00 158.95 348.75 368.05 348.11 285.61 220.10 162.69 121.23 91.50 70.04 54.34 33.88 0.00
3.25 20.32 222.01 454.06 470.05 447.47 351.41 283.28 227.22 169.32 127.81 97.83 75.90 47.32 27.36
3.75 27.05 295.57 553.89 557.95 543.23 467.85 352.17 280.17 225.42 170.15 130.24 101.05 62.99 36.43
Significant Wave Height (Hsig m)

4.25 34.75 378.26 651.04 675.34 607.63 542.34 452.34 337.44 269.11 218.55 167.29 129.79 80.91 46.79
4.75 43.40 462.09 764.90 780.12 720.56 581.38 506.60 421.51 336.15 265.18 208.97 162.13 101.07 58.45
5.25 53.02 540.87 882.10 873.13 800.09 677.84 526.13 459.68 386.21 310.76 248.16 198.06 123.47 71.40
5.75 63.60 628.27 1013.64 1014.18 869.35 731.05 604.59 525.38 463.28 351.36 285.88 231.09 148.11 85.65
6.25 75.15 700.46 1151.14 1156.46 990.70 824.70 714.31 550.42 487.24 415.12 337.76 273.03 174.98 101.19
6.75 87.65 769.33 1280.72 1268.36 1115.63 927.70 743.82 616.64 542.70 461.67 371.93 306.10 204.10 118.04
7.25 101.12 884.30 1438.06 1399.78 1284.45 1027.26 823.31 711.37 551.56 496.26 429.07 333.80 229.22 136.17
7.75 115.54 949.39 1562.78 1564.34 1434.34 1110.62 913.01 722.28 606.49 542.40 468.12 381.43 261.93 155.60
8.25 130.93 1075.84 1714.17 1703.32 1444.07 1257.94 989.96 818.48 687.28 538.97 492.92 432.23 285.64 176.32
8.75 147.29 1210.20 1854.07 1811.52 1586.00 1388.26 1061.09 886.35 735.78 584.22 531.01 464.71 321.32 198.34
9.25 164.60 1292.81 1963.71 1913.19 1666.77 1359.18 1160.88 965.36 765.18 652.89 593.44 481.59 340.04 221.66
9.75 182.87 1428.70 2086.10 1971.12 1762.75 1483.82 1289.46 1001.38 820.38 691.79 576.19 512.90 377.80 239.92
10.5 212.09 1606.36 2166.21 1999.15 1845.53 1593.59 1282.81 1109.86 929.98 743.33 644.59 594.85 438.16 268.05
11.5 254.41 1825.85 1866.23 1665.76 1894.27 1737.70 1477.92 1312.62 1007.71 861.95 738.56 600.43 465.24 321.54
12.5 300.58 2040.55 1569.07 1449.63 1658.27 1788.35 1604.74 1314.60 1154.26 950.23 779.96 709.40 527.64 360.30
13.5 350.60 2135.44 1344.10 1272.12 1468.83 1607.84 1687.46 1455.93 1328.33 1048.17 909.75 728.90 534.54 420.26
14.5 404.46 2235.75 1191.13 1132.20 1310.63 1605.11 1728.14 1571.18 1302.16 1169.89 956.40 814.77 596.00 436.58
15.5 462.18 2154.13 1048.72 1013.06 1172.76 1451.34 1547.39 1636.59 1416.77 1317.09 1038.21 931.02 681.05 489.96
16.5 523.73 1928.95 960.29 931.84 1070.50 1322.54 1587.80 1671.05 1522.08 1266.09 1157.95 961.72 739.36 533.61

Figure 2.15 Pelamis Performance Matrix (supplied by OPD)

Due to the Pelamis being an accumulator device, it is a more complex task to


calculate the proportion of energy it extracts from oncoming waves compared to a
terminator device. Using equation (2.5) a similar matrix was constructed containing
available power flux values in kW/m for the range of sea states. Then the values were
compared to see what the Pelamis power output was for an available power flux. A
maximum figure of 18.60 was calculated with the unit of Pelamis output in kW per
kW/m crest of available power. This figure would represent the maximum relative
efficiency of the Pelamis due to it extracting the highest amount of power given the

55
power available in the wave. All the other values were calculated as a percentage
relative to this maximum value to produce Figure 2.16 below.

tz (s) Efficiency % 75-100 50-75 25-50 0-25


Hsig (m) 3.5 4.5 5.5 6.5 7.5 8.5 9.5 10.5
0.25 0 0 0 0 0 0 0 0
0.75 0 0 100 97 73 50 0 0
1.25 0 50 100 97 73 50 33 22
1.75 0 50 100 97 73 50 33 22
2.25 0 50 100 93 70 50 33 22
2.75 0 50 90 81 66 48 33 22
3.25 6 50 84 74 61 42 30 22
3.75 6 50 77 66 55 42 28 20
4.25 6 50 70 62 48 38 28 19
4.75 6 49 65 55 46 33 25 19
5.25 6 47 53 45 39 31 22 17
5.75 6 45 44 38 33 28 21 16
6.25 6 43 38 32 28 24 21 14
6.75 6 39 32 27 24 21 18 14
7.25 6 34 28 24 20 18 16 14
7.75 6 30 24 21 18 16 14 12
8.25 6 26 22 18 16 14 12 11
8.75 6 23 19 16 14 12 11 10
9.25 6 21 17 15 13 11 10 9
9.75 6 19 15 13 11 10 9 8

Figure 2.16 Relative efficiency of Pelamis

It must be stressed that this represents relative output efficiency, not the true
efficiency defined as the percentage of energy extracted from incident waves. This
matrix indicates that the efficiency of the Pelamis reduces significantly as the level of
incident wave energy increases throughout most of its operating range. This is quite
unusual as most wave energy converters including OWC types (Beyene & Wilson
2005) are normally less efficient at part load. However this appears to be an
advantage of the Pelamis to be able to extract so much power from small seas,
because this represents the majority of total generation time, hence improving the
capacity factor.

In later sections of this report, wave data sets are analysed and provide an indication
of the occurrence of certain sea states. While it is true that wave conditions spend
most of the time with Hs between 1 and 5 metres (with a mode of 2 metres) the zero
crossing period is usually between 6.5 and 8.5 seconds. This suggests the Pelamis
should be designed for greater efficiency at these longer wave periods. OPD indicate
that there is some degree of tuning available with the Pelamis to optimise
performance for local conditions.

56
2.5 Project Discussion and Methodology

Most of section 2.1 was a research process to educate the author on the broader topic
of renewable energy, and was an ongoing task throughout the project. It was
necessary to view wave energy in context with all other forms of energy generation
and to see how the Pelamis competes with these technologies, wind turbines in
particular. It presents an up to date scenario of major engineering projects and
technologies. The section researching energy storage and the hydrogen economy was
found to be very relevant to the future use of renewable energy.

Section 2.2 contains most of the fundamentals of ocean waves and their energy
analysis, providing key equations and phenomena. This is critical to undertaking an
analysis of wave data and understanding how to analyse the wave resources specific
to southern Australia. The level of research can quickly become very difficult when
trying to research ocean wave phenomena, which is very complex and chaotic.
Initially it was intended to develop a mathematical model on MATLAB of wave
energy and Pelamis output South of American River, Kangaroo Island. After some
perseverance it became apparent just how enormous this task would be and with a
great margin for error and failure. It would not be an efficient use of time in meeting
the project goals and has already been done before by teams of mathematicians. So
the idea was abandoned and instead it was decided to perform a detailed statistical
analysis of wave rider buoy data and then create simple empirical models of Pelamis
performance at the buoy sites. Then it was discovered there was a student working
with Ocean Prospect already working specifically on creating a MCP model for waves
(see section 3.5).

Section 2.3 looks at the different wave energy converter designs to see how the
Pelamis compares and performs. From a commercial standpoint it certainly appears to
be the leading design currently. This section was an attempt to critique the Pelamis
and identify design flaws, but eventually it became accepted that it is a superior
design, but not impervious to improvements. A lot of specific questions to OPD
which might have led to design improvements could not be answered due to
commercial confidentiality. The task of this project was not to redesign the Pelamis or
improve its abilities as a wave energy converter, Wind Prospect was not asking for

57
this, so this was left to the engineers at OPD. So sections 2.3 and 2.4 became a means
to better understand how the technology works and provides only a brief summary of
the total amount of research that was undertaken.

One thing that was identified was a concern for the sheer mass of steel (originally
thought to be 700 tons) and how this might affect the embodied energy of the design.
After OPD declined to provide a lifecycle analysis and with some interest from Wind
Prospect, it was decided to do some research into embodied energy and energy
balances to make some calculations for the Pelamis. This led to the work in Chapter 6.

It became apparent that the originally intended search for Pelamis components was
not going to be possible because OPD would not provide the necessary information
and specifications due to confidentiality. Communication with Wind Prospect led to
realizing potential developments for Pelamis desalination. This occurred some time
after the literature review had been completed and a great deal of resource analysis
had been completed. Given the amount of time that was still available it was decided
that research in the area would be beneficial to Wind Prospect and the project. This
led to the work in section 5.2, where section 5.2.2 was a focused research task into
reverse osmosis technology to see how it could apply to the Pelamis.

The case studies also needed a great amount of broad research that could not have
been anticipated for in the literature review. Early work analysing the Kangaroo
Island resource led to the realization that there might be other better sites along the
southern Australian coast and that ideal conditions needed short distances from shore
to the 50 metre depth contour. This led to the bathymetric survey, which turned out to
be a very time consuming but valuable process.

A record was documented for all important meetings, particularly the early ones with
ETSA Utilities, Air Ride and a representative from OPD. Records of communications
(email in particular) and effective information management became very important to
the success of the project, particularly because it was research heavy for a mechanical
engineering final year project.

58
3 Southern Australian Wave Energy Resources
This chapter examines existing wave energy resources for southern Australia with the
aim of identifying locations best suited for the use of the Pelamis. Acquiring highly
detailed and accurate information of wave energy resources for just a single location
is a time consuming process and in many cases very expensive. Effort has been made
to make the best possible use of freely available information. A list of all the wave
information resources used in this chapter can be found in APPENDIX C. The most
in-depth analysis into wave energy is for a single location off Cape Du Couedic,
Kangaroo Island, in section 3.2. Many aspects of the study in section 3.2 will apply to
wave energy estimation and prediction at any coastal location. Much of the work
involved in this chapter meant finding only enough information in order to make
reasonable recommendations or provide directions on where to look next.

3.1 Australia's Long Term Wave Climate

Section 2.2.1 provides a description of Australia’s long term wave climate and the
processes that form these wave patterns. Figures 2.4, 2.5 and 2.6 give some indication
of the westerly directional climate and wave energy levels available for the southern
coasts of Australia. This section looks at the wave climate depicted by the long term
model called the ERA-40.

The ERA-40 Project was undertaken by the European Centre for Medium Range
Weather Forecasts (ECMWF 2005). It is a worldwide atmospheric reanalysis
constructed using wind data and covering a 40 year period after 1957. The ERA-40
model does not account for shallow water effects and therefore the ERA-40 ocean
wave data are only valid in deep water regions (>100m). Full data sets can be
purchased from the ECMWF. The Global Wave Climatology Atlas (Sterl et al. 2003)
is a report containing long term wave statistics derived from the ERA-40 analysis. It
is freely available for research purposes, however no data sets are available, only plots
of limited resolution. Some important plots have been extracted from this report and
will be summarized here.

59
Figure 3.1 Global ERA-40 plots for yearly mean Hs and Tz (Sterl et al. 2003)

60
The two contour plots shown in Figure 3.1 are derived from mean values over the
period shown on the top of each plot. Focusing on southern Australia, the long-term
mean Hs is 3 to 3.5m and mean Tz is 8 to 9 seconds. Using equation (2.5) these values
would equate to average annual power flux of between 36 and 55 kW/m. These values
appear to be substantially less than those shown in Figure 2.5 from Duckers (2004), a
widely used source of information. More detailed analysis in section 3.2.2 shows a
lower value still at about 32 kW/m for Cape Du Couedic, Kangaroo Island.

To get an understanding of the long term directional climate, some plots are available
from the Atlas displaying monthly mean wave directional arrows in APPENDIX E.
Again focusing on southern Australia, although the resolution is poor it is easy to see
a south westerly trend over the whole region. Looking at the plot for July and also the
last plot for June, there appears to be a tendency for a more direct westerly climate in
winter, when wind and wave energy levels are highest. The plots in APPENDIX E
also display monthly mean Hs, giving an indication of seasonal wave climate. Waves
during July are an order of 1 to 1.5 meters higher than in January, indicating much
higher mean wave energy levels. A clearer indication of the directional wave climate
arriving at South Australia’s coasts is given in section 3.2.3.

61
3.2 Study of Waves Arriving at Cape Du Couedic

In November 2000, the Bureau of Meteorology deployed the first wave rider buoy in
South Australian coastal waters (Watson 2001). It still remains moored in an 80m
depth of water located 4 nautical miles (~8km) west of Cape Du Couedic, on the
southwest extremity of Kangaroo Island. It is ideally located to measure unobstructed
long fetch swell from the Southern Ocean arriving at the centre of South Australia’s
coastline. A report was released by the Bureau of Meteorology (Watson 2001)
containing some preliminary analysis of the buoy data (March to August 2001) and
comparing it to wind data collected at Neptune Island. APPENDIX F contains
extracts from this report including the abstract and information about the
specification, deployment, location and operation of the buoy. For the purposes of this
project, 3½ years worth of data have been acquired from the Bureau of Meteorology
to study wave energy levels at Cape Du Couedic.

3.2.1 Data Description

The available buoy data were collected from 29th November 2000 through to the 28th
April 2004. The total number of data recordings (rows) taken during this period was
counted at just over 60,000. The Datawell Wave Rider Buoy at Cape Du Couedic (see
APPENDIX F) uses an accelerometer to measure time intervals between different
positions in the amplitude of the wave spectra. Effectively the buoy is able to record
information describing the wave heights and periods but not the direction of the
incident waves (more advanced buoys can measure direction).

DD/MM/YYYY HH:MN:SS Hs Hrms Hmax Tz Ts Tc THmax T02 Tp Hrms EPS


m m m s s s s s s m -
5/02/2002 20:54:24 2.62 1.84 4.21 6.57 8.82 3.8 11.03 6.43 12.01 1.96 0.84
5/02/2002 21:21:03 2.59 1.84 3.89 6.6 8.99 3.87 9.91 6.36 10.86 1.93 0.82
5/02/2002 21:47:43 2.48 1.72 4.16 6.09 8.41 3.95 9.83 6.3 11.69 1.88 0.82
5/02/2002 22:14:23 2.36 1.68 3.79 6.1 8.05 3.85 7 6.17 11.77 1.84 0.81
5/02/2002 22:41:03 2.49 1.73 4.25 6.1 8.13 3.94 8.84 6.23 10.56 1.89 0.81
5/02/2002 23:07:43 2.39 1.69 4.27 6.32 8.1 4.06 8.49 6.21 7.98 1.79 0.81
6/02/2002 0:01:07 2.41 1.68 3.97 6.44 9.03 3.99 9.54 6.56 11.81 1.83 0.84
6/02/2002 0:27:47 2.49 1.78 3.95 6.58 8.38 3.94 7.18 6.33 8.06 1.87 0.83
6/02/2002 0:54:27 2.33 1.63 4.31 6.18 8.09 3.89 9.22 6.38 8.87 1.79 0.82
6/02/2002 1:21:07 2.28 1.62 4.23 6.39 8.67 3.8 8.66 6.24 11.68 1.73 0.84

Figure 3.2 Sample of wave rider buoy data

62
A sample of the data when exported to Excel is shown in Figure 3.2. A description of
each of these data fields can be found in APPENDIX G, mainly related to how a
waverider buoy calculates its data. The two most important and useful values from
these data are those representing significant wave height (Hs) and zero crossing period
(Tz). Explanations of these values are provided at the end of section 2.2.2. These two
values are used primarily in the statistical analysis of wave data.

As can be seen in Figure 3.2, the average interval between data recordings from the
Cape Du Couedic buoy was 26:40 (mins:secs). However this was not always the case
and it has been discovered there are a few variations throughout the data. For the most
part, only the full 2001, 2002 & 2003 data will be used in the statistical analysis for
the sake of annual trending. Table 3.1 displays the total number of recordings for each
month so that discrepancies can be identified.

Table 3.1 Total number of recordings per month 2001-2003


Jan-01 3213 Jan-02 1590 Jan-03 1439
Feb-01 2430 Feb-02 1443 Feb-03 1289
Mar-01 1592 Mar-02 1563 Mar-03 1411
Apr-01 1536 Apr-02 272 Apr-03 1325
May-01 1588 May-02 0 May-03 1390
Jun-01 1317 Jun-02 1315 Jun-03 1307
Jul-01 1588 Jul-02 1151 Jul-03 1383
Aug-01 1583 Aug-02 1429 Aug-03 1388
Sep-01 1528 Sep-02 1334 Sep-03 1324
Oct-01 1578 Oct-02 1400 Oct-03 1415
Nov-01 1533 Nov-02 1391 Nov-03 1386
Dec-01 1588 Dec-02 1065 Dec-03 1433

It appears that the recording interval for most of January and February 2001 was
13:40 (mins:secs) until 2pm on 20th February 2001, when it changed to 26:40
(mins:secs). This accounts for the higher number of recordings in those months. Then
the interval remained 26:40 (mins:secs) until 2pm on April 7th 2002 when the
recordings stop. No data was recorded in May 2002 and then the recordings continue
at 2pm on the 1st of June 2002 with intervals of 30:00 (mins:secs). From there the
interval remains the same until the end of 2003, making 2003 the most complete and
consistent annual data set. Differences in the values of monthly totals for later in 2002
compared to 2003 can be explained by the observation that some of the data values
appear to be missing, probably removed due to random recording errors being

63
identified. All of these discrepancies will become more important for the statistical
analysis in the next section.

3.2.2 Statistical Analysis and Wave Climate Trends

All the graphs and tables in this analysis were generated using Microsoft Excel and
simple Visual Basic programming was used to rearrange and calculate data. Table 3.2
displays a summary of mean values for the years 2001, 2002 & 2003. These means
were taken for all the data values in the year and include minor errors due to the
discrepancies described earlier. Although the annual variation in mean significant
wave height between 2002 and 2003 looks small at 25 cm or 9%, it corresponds to
quite a large variation in the mean available power at 15% for those years. Note that
using equation (2.5) on the Hs and Tz mean values will not give the mean available
power values. Indeed the calculated values are significantly smaller. This can be
explained by the distribution of significant wave heights explored later. The averaged
values are simply the mean of the other 3 mean values, not means of all the data.

Table 3.2 Summary of Mean Values at CDC


Mean Value 2001 2002 2003 Averaged
Hs (metres) 2.54 2.79 2.60 2.65
Tz (seconds) 7.60 7.77 7.77 7.71
Available Power (kW/m) 31.15 36.82 30.00 32.66

For the record, the single largest wave recorded was Hmax = 14.84 metres at 2:30am
on the 28th June 2002 with a corresponding period of THmax = 11.99 seconds. Given
that values nearly as large were recorded either side of this one suggests that it might
be an accurate description of the true dimensions of the wave. If that is the case, using
equation (2.2), this wave would have a power flux of:
ρg 2 H 2T 1000 × 9.812 × 14.84 2 × 11.99
P(Wm −1 ) = = = 2528kWm −1 (3.3)
32π 32π
Assuming that the energy level was fairly consistent along the wave crest, had 100%
of the energy been extracted from about 600 metres crest length of this wave and
converted to electricity, then theoretically it could have momentarily powered South
Australia’s average load demand (1,500MW). Having been designed for the 100 year
storm wave, the Pelamis should have easily survived this wave.

64
Significant Wave Height Zero Crossing Period
4 9
3.75 8.75
3.5 8.5
3.25 8.25
3 8

seconds
metres

2.75 7.75
2.5 7.5
2.25 7.25
2 7
1.75 6.75
1.5 6.5
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Available Power Flux


55
50
45
40
35
kW/m

30
25
20
15
10
5
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Figure 3.3 Series of plots from combined monthly mean values

Data for each month of the years ’01, ’02 & ’03 were combined and then the means
were taken. Figure 3.3 displays all of these mean values for Hs, Tz and available
power flux, which was calculated using equation (2.5) for every data value. The
information in these graphs should by no means be used as an accurate prediction
tool; they merely indicate the trends from the 3 years available. There is solid
evidence of seasonal cycles. Wave periods are longer in the cooler months containing
more powerful long fetch swell and shorter in the calmer summer months. Wave
energy levels are closely related to significant wave height, high levels during the
winter months and then apparently peaking in September. Over the summer months
the mean power levels appear to remain level at just under 20kW/m. If many more
years of data were available, the available power curve would probably be smoother
and relate closely to mean seasonal wind velocities over the fetch to Kangaroo Island.
Care is needed not to read too much into Figure 3.3, as the data is highly variable as
seen in the following figures.

65
Significant Wave Height (m) Wave Period Tz (s)
9

0
May-01

May-02

May-03

May-04
Nov-00

Jan-01

Mar-01

Jun-01

Nov-01

Jan-02

Mar-02

Jun-02

Nov-02

Jan-03

Mar-03

Jun-03

Nov-03

Jan-04

Mar-04

Jun-04
Aug-00

Jul-01
Aug-01

Jul-02
Aug-02

Jul-03
Aug-03
Sep-00
Oct-00

Dec-00

Feb-01

Apr-01

Sep-01
Oct-01

Dec-01

Feb-02

Apr-02

Sep-02
Oct-02
Oct-02

Dec-02

Feb-03

Apr-03

Sep-03
Oct-03

Dec-03

Feb-04

Apr-04
Figure 3.4 Plot of monthly means for all 3½ years Hs and Tz data

Monthly means were taken for all of the 3½ years of data and a graph was generated
in Figure 3.4, with the missing values for May 2002 interpolated. A similar plot for all
of the daily means is very difficult to read due to the strongly variable nature of the
data. From this graph it is also easy to see seasonal trends but the variation in mean
monthly levels from year to year are now also clearly visible. The monthly mean
values of Hs do not drop below 2 metres but only exceed 3 metres during the winter
months and September. Mean values of Tz appear to roughly follow the same trend
patterns as Hs with shorter periods only occurring in the warmer months. The most
violent 3 month period occurred in the winter of 2002 and the summer of 2001/2002
was perhaps the most stable and calmest 3 month period. A comparison of these two
periods is shown in Figure 3.5 below, where means were taken for every day.

Hs (m) Tz (s) Hs (m) Tz (s)


12 12

10 10

8 8

6 6

4 4

2 2

0 0
Jun-02 Jul-02 Jul-02 Aug-02 Oct-01 Nov-01 Dec-01 Jan-02

Figure 3.5 Daily mean values of winter ’02 (left) and summer ’01/02 (right)

66
These graphs give an impression of how variable the data is from day to day even
though they represented the extremes of the monthly mean values. During the rougher
months of winter ’02 the values of Hs rarely dropped below 2 metres and peaks
followed variable cycles. It shows that one day during winter the mean Hs could be
around 5 metres, with Tz = 9s, and then only a couple of days later drop to 2.5 metres,
with Tz = 7s. A quick calculation suggests a daily mean wave energy resource could
fluctuate between 112 and 22 kW/m several times per month in winter. In the calmer
months of summer ’01/02 there appears to be a smaller variation and more
monochromatic cycles lasting about 5 days, corresponding to typical weather cycles.
In these months the energy levels frequently dropped below 15 kW/m. This is
important for addressing firm capacity. There appears to be a closer correlation
between Hs and Tz curves for daily means than monthly means and in particular when
values of Hs exceed 3 metres.

A much more useful way of illustrating the variation in wave data is to look at the
spectral distribution in a series of histograms. Comparing all three years of recordings
has it’s limitations due to the discrepancies listed in the previous section. The most
complete set of annual data is for 2003 and histograms for this year are displayed in
the following 3 figures. Then comparisons are made with histograms generated from
daily means of all ’01, ’02 & ’03 shown in the 3 figures after the 2003 histograms.

Although 2003 may not represent the same mean levels as the other years, from the
17,062 recordings used for each histogram there should be some clarity and accuracy
to the annual distribution trends. Figure 3.6 shows that the Hs distribution is
moderately skew to the right with the range not exceeding 6 meters overall. The mode
is just under 2 metres or about 2/3 of one stdev left of the mean. Figure 3.7 for Tz is
much closer to a normal distribution making for easy prediction, but slightly skew to
the right also. The power histogram in Figure 3.8 is very much skew to the right; an
exaggeration of the Hs distribution. The curve is quite smooth and it may be possible
to fit an equation to different parts of the line. The mode is about 1/3 of the mean and
to its right the curve tends exponentially to the horizontal axis. A number of outliers
existed far beyond 150kW/m, describing the infrequency of powerful storm or ‘freak’
waves.

67
Significant Wave Height 2003
1700

frequency

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 7 7.5 8
metres

Figure 3.6 2003 Histogram for Hs: mean = 2.60, stdev = 1.05

Zero crossing Period 2003


2500
frequency

0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
seconds

Figure 3.7 2003 Histogram for Tz: mean = 7.77, stdev = 1.45

Available Power Flux 2003


3000
frequency

0
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150
kW/m

Figure 3.8 2003 Histogram for Power flux: mean = 30.00, stdev = 31.95

68
Due to the discrepancies in the data for 2001 and 2002, histograms taken for all 3
years of recorded data would contain a number of errors. Taking mean daily values
should go some way to eliminating these errors apart from the hole in the data from
7th April to 1st June 2002. There is also a sacrifice of resolution to the distribution with
a total sample size of 1,030 values compared to the 17,062 in 2003. The following
three figures confirm that there is no significant alteration in trend to 2003, but they
represent 3 years instead of one. The 2003 histograms could be considered trend
models through which the next three histograms can be compared.

Figure 3.9 shows positive resemblance to Figure 3.6 and it can be said that Hs very
rarely falls below 1 metre or exceeds 5.5 metres. The Tz histogram in Figure 3.10
again is quite close to a normal distribution. Using the mean of 7.76 and the stdev
from Figure 3.7 (1.45) an approximation can be made that Tz spends 68% (±2σ) of the
time between 4.86 and 10.66 seconds. By observation also, the value of Tz almost
never falls below 4 seconds or goes above 12 seconds. There is a bit more noise in
Figure 3.11 than for the 2003 plot, but again the distribution is the same. For half of
the time the available power flux at Cape Du Couedic is between 13 and 44 kW/m
(interquartile range) and spends half the time either side of 24kW/m (median).

Another way to visualize the spread of the data is to produce a scatter diagram. Figure
3.12 is a plot of every recording during 2003, the most complete data set for Cape Du
Couedic. The histograms are most effective at illustrating the distribution of a
particular value and its central tendency, but the scatter diagram shows what wave
dimensions are likely. It can be seen that waves with Hs below 1 metre do not occur
when Tz is below 4 seconds and become very unlikely when Tz is higher than 7
seconds. This information is useful in determining the likelihood of the Pelamis
generating zero or negligible amounts of power, investigated in section 4.1.3. A linear
line has been added between Hs=2, Tz=5 and Hs=7, Tz=10. It could be claimed that
waves of dimensions above this line are very rare occurrences. This region
corresponds to the majority of the instances where a Pelamis reaches it 750kW limit
as highlighted in the performance matrix in Figure 2.15.

69
Significant Wave Height - 2001/2/3 daily means
150

frequency

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 7 7.5 8
metres

Figure 3.9 Overall histogram for Hs: mean = 2.68, stdev = 1.00

Zero crossing Period - 2001/2/3 daily means


180
frequency

0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
seconds

Figure 3.10 Overall histogram for Tz: mean = 7.76, stdev = 1.26

Available Power Flux - 2001/2/3 daily means


70
frequency

0
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150
kW/m

Figure 3.11 Overall histogram for Power Flux: mean = 33.66, stdev = 29.42

70
Scatter Diagram 2003
7

5
Hs (metres)

0
3 4 5 6 7 8 9 10 11 12 13
Tz (seconds)

Figure 3.12 Scatter Diagram of all year 2003 recordings

In Figure 3.13 an attempt was made to model the distribution for a 24 hour period,
although the results may not provide much of an accurate indication. It was decided to
take mean values of each hour from the whole data set (≈60,000 or 2,500 values/hour)
to see what kind of spread occurred. As anticipated, the data looks irregular and there
are very small variations overall. A trend line was fitted to each plot, a 3rd degree poly
to Hs and Power curves and a 2 degree poly to the Tz curve. The graphs show a
cyclical tendency for wave energy levels to be slightly higher around sunset than
sunrise. However the variation in mean values is probably too small to be significant
compared to random variability. During steady or calmer periods, these trends could
be more visible or magnified. The main contributor to a 24 hourly trend would be
daily wind cycles. These kinds of cycles may be more predictable on a long straight
section of coastline (unlike Cape Du Couedic) where wind rushes on and off the land
in more measurable cycles and often normal to the coast.

71
24 Hour Mean Significant Wave Height 24 Hour Mean Zero crossing Period
2.58
7.7
2.56
7.65
2.54

seconds
metres

7.6
2.52
7.55
2.5
7.5
2.48
7.45
2.46
0 4 8 12 16 20 24
0 4 8 12 16 20 24
hour
hour

24 Hour Mean Wave Power Flux


42.5
42
41.5
41
kW/m

40.5
40
39.5
39
38.5
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
hour

Figure 3.13 24 hour distributions from mean hour values

A more useful way of measuring the variation of the data is to calculate the day to day
variations by taking the RMS. Again the 2003 data was used and the following mean
variations ∆ were found: ∆Hs= 0.618 metres, ∆Tz= 0.972 seconds and the mean day to
day power variation was 16.57 kW/m. These values would correspond to what should
be an expected variation in daily mean values at about 50% of the annual mean value.
Histograms were generated (not reproduced here) to see the distributions of these
variations and a skew to the right profile was generated in each case. Similar to the
available power distribution Figure 3.8 except the mode is much closer to the zero
variation value.

72
3.2.3 Wind to Wave Correlations and Directional Climate

The preliminary report released by the Bureau of Meteorology analysing the Cape Du
Couedic buoy data (Watson 2001) provides a very useful indication of the directional
climate of long swell waves arriving at South Australia. The summary of this report in
APPENDIX F gives the results of this analysis using data from March to August
2001. The Bureau of Meteorology makes great use of this buoy for coastal forecasting
and no doubt a lot more analysis has been done since. The important parts of this
analysis by Watson (2001) will be explained here.

The winds passing by Neptune Island would only differ slightly to Cape Du Couedic
even though it is 100kms away. Neptune Island just happens to be the nearest
automatic weather station but its data could be considered as local wind to Cape Du
Couedic. Figure 3.14 shows mean wind speed and wave height collected over the
month of August. It shows that the variability of wind speed is significantly greater
than that of wave height. Over the 5 months there was always a positive correlation
(R=0.47) between wind speed and wave height regardless of wind direction. Another
important observation is that there are a number of instances when wind energy is low
or zero and wave energy is strong. There are also clear instances where wind speed
would increase and wave height would decline. Correlations were much stronger for
winds from certain directions, namely the westerly and southerly sectors. The lag
between wind speed and wave height also varies depending on direction.

Figure 3.15 shows wave height compared to wind direction during August. It was
noted by Watson (2001) that an abrupt increase in wave height resulted from winds
shifting from north (45° to 315°) to the west (315° to 225°) that usually occurs with
the passage of a cold front. Negative correlations were shown for north and easterly
winds, indicating that energy can be taken from long swell waves by local winds.
Figure 3.16 provides a convenient display of wind speed and direction correlation
over the 5 months. From this graph it can be postulated that the directional wave
climate at Cape Du Couedic, and the rest of southern Australia, is westerly to south
westerly particularly during the strong winds occurring in winter months. Obtaining
figures like this has its uses in short term wave energy forecasting for any location
and in gaining an idea of long swell directions if directional data isn’t available.

73
8 40
Mean Wave = 3.42 Sig Wave
7 Mean Speed =19.39 Mean Wind 35
R = 0.47
6 30

Wind Speed (kts)


5 25
Height (m)
4 20

3 15

2 10

1 5

0 0
0:00
0:30
1:30
1:00
2:00
3:00
2:30
3:30
4:00
5:00
4:30
5:30
6:30
6:00
7:00
10:00
8:00
7:30
11:00
10:30
8:30
11:30
9:30
9:00
12:00
13:00
12:30
13:30
14:30
14:00
15:00
16:00
15:30
16:30
17:30
17:00
18:00
18:30
19:30
19:00
20:00
21:00
20:30
21:30
22:30
22:00
23:00
0:30
0:00
1:00
2:00
1:30
2:30
3:00
4:00
3:30
4:30
5:30
5:00
6:00
7:00
6:30
10:00
7:30
10:30
8:30
8:00
11:00
9:00
12:00
11:30
9:30
12:30
13:30
13:00
14:00
15:00
14:30
15:30
16:30
16:00
17:00
17:30
18:30
18:00
19:00
20:00
19:30
20:30
21:30
21:00
22:00
23:00
22:30
0:00
1:00
0:30
1:30
2:00
3:00
2:30
3:30
4:30
4:00
5:00
6:00
5:30
6:30
7:30
7:00
10:00
8:00
11:00
10:30
9:00
8:30
11:30
9:30
12:30
12:00
13:00
14:00
13:30
14:30
15:30
15:00
16:00
16:30
17:30
17:00
18:00
19:00
18:30
19:30
20:30
20:00
21:00
22:00
21:30
22:30
23:00
0:00
0:30
1:00
2:00
1:30
2:30
3:30
3:00
4:00
5:00
4:30
5:30
6:30
6:00
7:00
10:00
8:00
7:30
10:30
8:30
11:30
11:00
9:00
12:00
9:30
13:00
12:30
13:30
14:30
14:00
15:00
15:30
16:30
16:00
17:00
18:00
17:30
18:30
19:30
19:00
20:00
21:00
20:30
21:30
22:30
22:00
23:00
23:30
1:30
0:00
1:00
2:00
1:30
2:30
3:30
3:00
4:00
5:00
4:30
5:30
6:30
6:00
7:00
10:00
7:30
10:30
8:30
8:00
11:30
11:00
9:00
12:00
9:30
13:00
12:30
13:30
14:00
15:00
14:30
15:30
16:30
16:00
17:00
18:00
17:30
18:30
19:30
19:00
20:00
21:00
20:30
21:30
22:00
23:00
22:30
0:00
1:00
0:30
1:30
2:30
2:00
3:00
4:00
3:30
4:30
5:30
5:00
6:00
6:30
7:30
7:00
10:30
10:00
8:00
11:00
9:00
8:30
12:00
11:30
9:30
12:30
13:00
14:00
13:30
14:30
15:30
15:00
16:00
17:00
16:30
17:30
18:30
18:00
19:00
20:00
19:30
20:30
21:00
22:00
21:30
22:30
23:00
0:00
0:30
1:30
1:00
2:00
3:00
2:30
3:30
4:30
4:00
5:00
5:30
10:00
7:00
6:00
10:30
7:30
11:30
11:00
8:30
8:00
12:00
9:00
12:30
9:30
13:30
13:00
14:00
15:00
14:30
15:30
16:30
16:00
17:00
18:00
17:30
18:30
19:30
19:00
20:00
20:30
21:30
21:00
22:00
23:00
22:30
0:00
1:00
0:30
1:30
2:30
2:00
3:00
4:00
3:30
4:30
5:00
6:00
5:30
6:30
7:30
7:00
10:30
10:00
8:00
11:00
9:00
8:30
11:30
9:30
12:30
12:00
13:00
14:00
13:30
14:30
15:30
15:00
16:00
17:00
16:30
17:30
18:30
18:00
19:00
19:30
20:30
20:00
21:00
22:00
21:30
22:30
23:00
0:00
0:30
1:30
1:00
2:00
3:00
2:30
3:30
4:00
5:30
4:30
6:00
7:00
6:30
10:00
7:30
10:30
8:30
8:00
11:30
11:00
9:00
12:00
9:30
12:30
13:30
13:00
14:00
15:00
14:30
15:30
16:30
16:00
17:00
18:00
17:30
18:30
19:00
20:00
19:30
20:30
21:30
21:00
22:00
23:00
22:30
23:30
0:30
0:00
1:00
2:00
1:30
2:30
3:00
4:00
3:30
4:30
5:30
5:00
6:00
7:00
6:30
10:00
7:30
10:30
8:30
8:00
11:00
9:00
12:00
11:30
9:30
12:30
13:30
13:00
14:00
15:00
14:30
15:30
16:30
16:00
17:00
17:30
18:30
18:00
19:00
20:00
19:30
20:30
21:30
21:00
22:00
23:00
22:30
0:00
1:00
0:30
1:30
2:00
3:00
2:30
3:30
4:30
4:00
5:00
6:00
5:30
6:30
7:30
7:00
10:00
8:00
11:00
10:30
9:00
8:30
11:30
9:30
12:30
12:00
13:00
14:00
13:30
14:30
15:30
15:00
16:00
16:30
17:30
17:00
18:00
19:00
18:30
19:30
20:30
20:00
21:00
22:00
21:30
22:30
23:00
0:00
0:30
1:00
2:00
1:30
2:30
3:30
3:00
4:00
5:00
4:30
5:30
6:30
6:00
10:00
7:00
10:30
8:00
7:30
11:30
11:00
8:30
12:00
9:00
13:00
12:30
9:30
13:30
14:30
14:00
15:00
15:30
16:30
16:00
17:00
18:00
17:30
18:30
19:30
19:00
20:00
21:00
20:30
21:30
22:30
22:00
23:00
0:00
1:00
0:30
1:30
2:30
2:00
3:00
4:00
3:30
4:30
5:30
5:00
6:00
7:00
6:30
7:30
10:30
10:00
8:00
11:00
9:00
8:30
12:00
11:30
9:30
12:30
13:30
13:00
14:00
14:30
15:30
15:00
16:00
17:00
16:30
17:30
18:30
18:00
19:00
20:00
19:30
20:30
21:30
21:00
22:00
22:30
23:00
0:00
0:30
1:30
1:00
2:00
3:00
2:30
3:30
4:30
4:00
5:00
6:00
5:30
6:30
7:00
10:00
8:00
7:30
11:00
10:30
8:30
11:30
9:30
9:00
12:30
12:00
13:00
13:30
14:30
14:00
15:00
16:00
15:30
16:30
17:30
17:00
18:00
19:00
18:30
19:30
20:30
20:00
21:00
21:30
22:30
22:00
23:00
0:30
0:00
1:00
2:00
1:30
2:30
3:30
3:00
4:00
5:00
4:30
5:30
6:00
7:00
6:30
10:00
7:30
10:30
8:30
8:00
11:30
11:00
9:00
12:00
9:30
12:30
13:30
13:00
14:00
15:00
14:30
15:30
16:30
16:00
17:00
18:00
17:30
18:30
19:30
19:00
20:00
20:30
21:30
21:00
22:00
23:00
22:30
23:30
0:30
0:00
1:00
2:00
1:30
2:30
3:30
3:00
4:00
4:30
5:30
5:00
6:00
7:00
6:30
10:00
7:30
10:30
8:30
8:00
11:00
9:00
12:00
11:30
9:30
12:30
13:30
13:00
14:00
15:00
14:30
15:30
16:30
16:00
17:00
18:00
17:30
18:30
19:00
20:00
19:30
20:30
21:30
20:00
22:00
23:00
22:30
0:00
1:00
0:30
1:30
2:30
2:00
3:00
3:30
4:30
4:00
5:00
6:00
5:30
6:30
7:30
7:00
10:30
10:00
8:00
11:00
9:00
8:30
11:30
9:30
12:30
12:00
13:00
14:00
13:30
14:30
15:30
15:00
16:00
17:00
16:30
17:30
18:00
19:00
18:30
19:30
20:30
20:00
20:00
22:00
21:30
22:30
23:30
23:00
0:00
1:00
0:30
1:30
2:00
3:00
2:30
3:30
4:30
4:00
5:00
6:00
5:30
6:30
7:30
7:00
10:00
8:00
11:00
10:30
9:00
8:30
11:30
9:30
12:30
12:00
13:00
14:00
13:30
14:30
15:30
15:00
16:00
16:30
17:30
17:00
18:00
19:00
18:30
19:30
20:30
20:00
21:00
22:00
21:30
22:30
23:00
0:00
0:30
1:00
2:00
1:30
2:30
3:30
3:00
4:00
5:00
4:30
5:30
6:30
6:00
7:00
10:00
8:00
7:30
10:30
8:30
11:30
11:00
9:00
12:00
9:30
13:00
12:30
13:30
14:30
14:00
15:00
15:30
16:30
16:00
17:00
18:00
17:30
18:30
19:30
19:00
20:00
21:00
20:30
21:30
22:30
22:00
23:00
0:00
1:00
0:30
1:30
2:30
2:00
3:00
4:00
3:30
4:30
5:30
5:00
6:00
7:00
6:30
7:30
10:30
10:00
8:00
11:00
9:00
8:30
12:00
11:30
9:30
12:30
13:30
13:00
14:00
14:30
15:30
15:00
16:00
17:00
16:30
17:30
18:30
18:00
19:00
20:00
19:30
20:30
21:30
21:00
22:00
22:30
23:00
0:00
0:30
1:30
1:00
2:00
3:00
2:30
3:30
4:30
4:00
5:00
6:00
5:30
6:30
7:00
10:00
8:00
7:30
11:00
10:30
8:30
11:30
9:30
9:00
12:30
12:00
13:00
13:30
14:30
14:00
15:00
16:00
15:30
16:30
17:30
17:00
18:00
19:00
18:30
19:30
20:30
20:00
21:00
21:30
22:30
22:00
23:00
0:30
0:00
1:00
2:00
1:30
2:30
3:30
3:00
4:00
5:00
4:30
5:30
6:00
10:00
7:00
6:30
10:30
7:30
11:30
11:00
8:30
8:00
12:00
9:00
12:30
9:30
13:30
13:00
14:00
15:00
14:30
15:30
16:30
16:00
17:00
18:00
17:30
18:30
19:30
19:00
20:00
20:30
21:30
21:00
22:00
23:00
22:30
0:00
1:00
0:30
1:30
2:30
2:00
3:00
4:00
3:30
4:30
5:00
6:00
5:30
6:30
7:30
7:00
10:30
10:00
8:00
11:00
9:00
8:30
11:30
9:30
12:30
12:00
13:00
14:00
13:30
14:30
15:30
15:00
16:00
17:00
16:30
17:30
18:30
18:00
19:00
19:30
20:30
20:00
21:00
22:00
21:30
22:30
23:00
0:00
0:30
1:30
1:00
2:00
3:00
2:30
3:30
4:00
5:00
4:30
5:30
6:30
6:00
7:00
10:00
8:00
7:30
11:00
10:30
8:30
11:30
9:30
9:00
12:00
13:00
12:30
13:30
14:30
14:00
15:00
16:00
15:30
16:30
17:30
17:00
18:00
18:30
19:30
19:00
20:00
21:00
20:30
21:30
22:30
22:00
23:00
0:30
0:00
1:00
2:00
1:30
2:30
3:00
4:00
3:30
4:30
5:30
5:00
6:00
7:00
6:30
10:00
7:30
10:30
8:30
8:00
11:00
9:00
12:00
11:30
9:30
12:30
13:30
13:00
14:00
15:00
14:30
15:30
16:30
16:00
17:00
17:30
18:30
18:00
19:00
20:00
19:30
20:30
21:30
21:00
22:00
23:00
22:30
0:00
1:00
0:30
1:30
2:00
3:00
2:30
3:30
4:30
4:00
5:00
6:00
5:30
6:30
7:30
7:00
10:00
8:00
11:00
10:30
9:00
8:30
11:30
12:30
12:00
9:30
13:00
14:00
13:30
14:30
15:30
15:00
16:00
16:30
17:30
17:00
18:00
19:00
18:30
19:30
20:30
20:00
21:00
22:00
21:30
22:30
23:00
0:00
0:30
1:00
2:00
1:30
2:30
3:30
3:00
4:00
5:00
4:30
5:30
6:30
6:00
10:00
7:00
10:30
8:00
7:30
11:30
11:00
8:30
12:00
9:00
13:00
12:30
9:30
13:30
14:30
14:00
15:00
15:30
16:30
16:00
17:00
18:00
17:30
18:30
19:30
19:00
20:00
21:00
20:30
21:30
22:30
22:00
23:00
0:00
1:00
0:30
1:30
2:30
2:00
3:00
4:00
3:30
4:30
5:30
5:00
6:00
7:00
6:30
7:30
10:30
10:00
8:00
11:00
9:00
8:30
12:00
11:30
9:30
12:30
13:30
12:00
14:00
14:30
15:30
15:00
16:00
17:00
16:30
17:30
18:30
18:00
19:00
20:00
19:30
20:30
21:30
21:00
22:00
22:30
23:00
0:00
0:30
1:30
1:00
2:00
3:00
2:30
3:30
4:30
4:00
5:00
6:00
5:30
6:30
7:00
10:00
8:00
7:30
11:00
10:30
11:30
8:30
12:30
12:00
9:30
9:00
13:00
13:30
14:30
14:00
15:00
16:00
15:30
16:30
17:30
17:00
18:00
19:00
18:30
19:30
20:30
20:00
21:00
21:30
22:30
22:00
23:00
0:30
0:00
1:00
2:00
1:30
2:30
3:30
3:00
4:00
5:00
4:30
5:30
6:00
7:00
6:30
10:00
7:30
10:30
8:30
8:00
11:30
11:00
9:00
12:00
9:30
12:30
13:30
13:00
14:00
15:00
14:30
15:30
16:30
16:00
17:00
18:00
17:30
18:30
19:30
19:00
20:00
20:30
21:30
21:00
22:00
23:00
22:30
0:00
1:00
0:30
1:30
2:30
2:00
3:00
4:00
3:30
4:30
5:00
6:00
5:30
6:30
7:30
7:00
10:30
10:00
8:00
11:00
11:30
9:00
8:30
12:30
12:00
9:30
13:00
14:00
13:30
14:30
15:30
15:00
16:00
17:00
16:30
17:30
18:30
18:00
19:00
19:30
20:30
20:00
21:00
22:00
21:30
22:30
23:00

01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
Date

Figure 3.14 Hs and wind speed against time (Watson 2001)

8 360
Sig Wave
Wind Direction
7 315

6 270

Wind Direction (T)


5 225
Height (m)

4 180

3 135

2 090

1 045

0 000
0:00
0:30
1:30
1:00
2:00
3:00
2:30
3:30
4:30
4:00
5:00
6:00
5:30
6:30
10:00
7:00
11:00
10:30
8:00
7:30
11:30
8:30
12:30
12:00
9:30
9:00
13:00
14:00
13:30
14:30
15:30
15:00
16:00
17:00
16:30
17:30
18:30
18:00
19:00
20:00
19:30
20:30
21:00
22:00
21:30
22:30
23:00
0:00
0:30
1:30
1:00
2:00
3:00
2:30
3:30
4:30
4:00
5:00
6:00
5:30
6:30
10:30
10:00
7:30
7:00
11:00
8:00
11:30
9:00
8:30
12:30
12:00
9:30
13:00
14:00
13:30
14:30
15:30
15:00
16:00
17:00
16:30
17:30
18:30
18:00
19:00
20:00
19:30
20:30
21:30
21:00
22:00
23:00
22:30
0:00
1:00
0:30
1:30
2:00
3:00
2:30
3:30
4:30
4:00
5:00
6:00
5:30
6:30
10:30
10:00
7:30
7:00
11:00
8:00
12:00
11:30
9:00
8:30
12:30
9:30
13:30
13:00
14:00
15:00
14:30
15:30
16:00
17:00
16:30
17:30
18:30
18:00
19:00
20:00
19:30
20:30
21:30
21:00
22:00
23:00
22:30
0:00
1:00
0:30
1:30
2:30
2:00
3:00
4:00
3:30
4:30
5:30
5:00
6:00
6:30
10:30
10:00
7:30
7:00
11:00
8:00
12:00
11:30
9:00
8:30
12:30
9:30
13:30
13:00
14:00
15:00
14:30
15:30
16:30
16:00
17:00
18:00
17:30
18:30
19:30
19:00
20:00
20:30
21:30
21:00
22:00
23:00
22:30
23:30
1:30
0:00
1:00
2:00
1:30
2:30
3:30
3:00
4:00
5:00
4:30
5:30
6:30
6:00
10:00
7:00
10:30
8:00
7:30
11:30
11:00
8:30
12:00
9:30
9:00
13:00
12:30
13:30
14:30
14:00
15:00
16:00
15:30
16:30
17:30
17:00
18:00
19:00
18:30
19:30
20:30
20:00
21:00
22:00
21:30
22:30
23:00
0:00
0:30
1:00
2:00
1:30
2:30
3:30
3:00
4:00
5:00
4:30
5:30
6:30
6:00
10:00
7:00
11:00
10:30
8:00
7:30
11:30
8:30
12:30
12:00
9:30
9:00
13:00
14:00
13:30
14:30
15:00
16:00
15:30
16:30
17:30
17:00
18:00
19:00
18:30
19:30
20:30
20:00
21:00
22:00
21:30
22:30
23:00
0:00
0:30
1:30
1:00
2:00
3:00
2:30
3:30
4:30
4:00
5:00
5:30
10:00
7:00
6:00
10:30
7:30
11:30
11:00
8:30
8:00
12:00
9:00
13:00
12:30
9:30
13:30
14:30
14:00
15:00
16:00
15:30
16:30
17:30
17:00
18:00
19:00
18:30
19:30
20:00
21:00
20:30
21:30
22:30
22:00
23:00
0:30
0:00
1:00
2:00
1:30
2:30
3:30
3:00
4:00
5:00
4:30
5:30
6:30
6:00
10:00
7:00
10:30
8:00
7:30
11:30
11:00
8:30
12:00
9:30
9:00
13:00
12:30
13:30
14:30
14:00
15:00
16:00
15:30
16:30
17:30
17:00
18:00
19:00
18:30
19:30
20:30
20:00
21:00
22:00
21:30
22:30
23:00
0:00
0:30
1:00
2:00
1:30
2:30
3:30
3:00
4:00
5:30
4:30
6:00
10:00
7:00
6:30
10:30
7:30
11:30
11:00
8:30
8:00
12:00
9:00
13:00
12:30
9:30
13:30
14:30
14:00
15:00
15:30
16:30
16:00
17:00
18:00
17:30
18:30
19:30
19:00
20:00
21:00
20:30
21:30
22:30
22:00
23:00
23:30
0:00
0:30
1:30
1:00
2:00
3:00
2:30
3:30
4:30
4:00
5:00
5:30
6:30
6:00
10:00
7:00
11:00
10:30
8:00
7:30
11:30
8:30
12:30
12:00
9:30
9:00
13:00
14:00
13:30
14:30
15:30
15:00
16:00
17:00
16:30
17:30
18:30
18:00
19:00
19:30
20:30
20:00
21:00
22:00
21:30
22:30
23:00
0:00
0:30
1:30
1:00
2:00
3:00
2:30
3:30
4:30
4:00
5:00
6:00
5:30
6:30
10:00
7:30
7:00
11:00
10:30
8:00
11:30
9:00
8:30
12:30
12:00
9:30
13:00
14:00
13:30
14:30
15:30
15:00
16:00
17:00
16:30
17:30
18:30
18:00
19:00
20:00
19:30
20:30
21:30
21:00
22:00
23:00
22:30
0:00
0:30
1:30
1:00
2:00
3:00
2:30
3:30
4:30
4:00
5:00
6:00
5:30
6:30
10:30
10:00
7:30
7:00
11:00
8:00
12:00
11:30
9:00
8:30
12:30
9:30
13:30
13:00
14:00
14:30
15:30
15:00
16:00
17:00
16:30
17:30
18:30
18:00
19:00
20:00
19:30
20:30
21:30
21:00
22:00
23:00
22:30
0:00
1:00
0:30
1:30
2:30
2:00
3:00
4:00
3:30
4:30
5:00
6:00
5:30
6:30
10:30
10:00
7:30
7:00
11:00
8:00
12:00
11:30
9:00
8:30
12:30
9:30
13:30
13:00
14:00
15:00
14:30
15:30
16:30
16:00
17:00
18:00
17:30
18:30
19:00
20:00
19:30
20:30
21:30
21:00
22:00
23:00
22:30
0:00
1:00
0:30
1:30
2:30
2:00
3:00
4:00
3:30
4:30
5:30
5:00
6:00
7:00
6:30
10:30
10:00
7:30
11:00
8:30
8:00
12:00
11:30
9:00
12:30
9:30
13:30
13:00
14:00
15:00
14:30
15:30
16:30
16:00
17:00
18:00
17:30
18:30
19:30
19:00
20:00
21:00
20:30
21:30
22:30
22:00
23:00
0:00
1:00
0:30
1:30
2:30
2:00
3:00
4:00
3:30
4:30
5:30
5:00
6:00
10:00
7:00
6:30
10:30
7:30
11:30
11:00
8:30
8:00
12:00
9:00
13:00
12:30
9:30
13:30
14:00
15:00
14:30
15:30
16:30
16:00
17:00
18:00
17:30
18:30
19:30
19:00
20:00
21:00
20:30
21:30
22:30
22:00
23:00
23:30
0:00
0:30
1:30
1:00
2:00
3:00
2:30
3:30
4:00
5:00
4:30
5:30
6:30
6:00
10:00
7:00
11:00
10:30
8:00
7:30
11:30
8:30
12:30
12:00
9:30
9:00
13:00
14:00
13:30
14:30
15:30
15:00
16:00
17:00
16:30
17:30
18:30
18:00
19:00
19:30
20:30
20:00
20:00
22:00
21:30
22:30
23:00
0:00
0:30
1:30
1:00
2:00
3:00
2:30
3:30
4:30
4:00
5:00
6:00
5:30
6:30
10:00
7:30
7:00
11:00
10:30
8:00
11:30
9:00
8:30
12:30
12:00
9:30
13:00
14:00
13:30
14:30
15:30
15:00
16:00
17:00
16:30
17:30
18:30
18:00
19:00
20:00
19:30
20:30
21:30
20:00
22:00
23:00
22:30
23:30
0:00
1:00
0:30
1:30
2:30
2:00
3:00
4:00
3:30
4:30
5:30
5:00
6:00
10:00
7:00
6:30
10:30
7:30
11:30
11:00
8:30
8:00
12:00
9:00
13:00
12:30
9:30
13:30
14:00
15:00
14:30
15:30
16:30
16:00
17:00
18:00
17:30
18:30
19:30
19:00
20:00
21:00
20:30
21:30
22:30
22:00
23:00
0:30
0:00
1:00
2:00
1:30
2:30
3:30
3:00
4:00
4:30
5:30
5:00
6:00
10:00
7:00
6:30
10:30
7:30
11:30
11:00
8:30
8:00
12:00
9:00
13:00
12:30
9:30
13:30
14:30
14:00
15:00
16:00
15:30
16:30
17:30
17:00
18:00
18:30
19:30
19:00
20:00
21:00
20:30
21:30
22:30
22:00
23:00
0:30
0:00
1:00
2:00
1:30
2:30
3:30
3:00
4:00
5:00
4:30
5:30
6:30
6:00
10:00
7:00
10:30
8:00
7:30
11:30
11:00
8:30
12:00
9:00
13:00
12:30
9:30
13:30
14:30
14:00
15:00
16:00
15:30
16:30
17:30
17:00
18:00
19:00
18:30
19:30
20:30
20:00
21:00
22:00
21:30
22:30
23:00
0:30
0:00
1:00
2:00
1:30
2:30
3:30
3:00
4:00
5:00
4:30
5:30
6:30
6:00
10:00
7:00
11:00
10:30
8:00
7:30
11:30
8:30
12:30
12:00
9:30
9:00
13:00
13:30
14:30
14:00
15:00
16:00
15:30
16:30
17:30
17:00
18:00
19:00
18:30
19:30
20:30
20:00
21:00
22:00
21:30
22:30
23:00
0:00
0:30
1:30
1:00
2:00
3:00
2:30
3:30
4:00
5:00
4:30
5:30
6:30
6:00
10:00
7:00
11:00
10:30
8:00
7:30
11:30
8:30
12:30
12:00
9:30
9:00
13:00
14:00
13:30
14:30
15:30
15:00
16:00
17:00
16:30
17:30
18:00
19:00
18:30
19:30
20:30
20:00
21:00
22:00
21:30
22:30
23:00
0:00
0:30
1:30
1:00
2:00
3:00
2:30
3:30
4:30
4:00
5:00
6:00
5:30
6:30
10:00
7:30
7:00
11:00
10:30
8:00
11:30
8:30
12:30
12:00
9:30
9:00
13:00
14:00
13:30
14:30
15:30
15:00
16:00
17:00
16:30
17:30
18:30
18:00
19:00
20:00
19:30
20:30
21:30
21:00
22:00
22:30
23:00
0:00
0:30
1:30
1:00
2:00
3:00
2:30
3:30
4:30
4:00
5:00
6:00
5:30
6:30
10:30
10:00
7:30
7:00
11:00
8:00
12:00
11:30
9:00
8:30
12:30
9:30
13:00
14:00
13:30
14:30
15:30
15:00
16:00
17:00
16:30
17:30
18:30
18:00
19:00
20:00
19:30
20:30
21:30
21:00
22:00
23:00
22:30
0:00
1:00
0:30
1:30
2:30
2:00
3:00
3:30
4:30
4:00
5:00
6:00
5:30
6:30
10:30
10:00
7:30
7:00
11:00
8:00
12:00
11:30
9:00
8:30
12:30
9:30
13:30
13:00
14:00
15:00
14:30
15:30
16:30
16:00
17:00
17:30
18:30
18:00
19:00
20:00
19:30
20:30
21:30
21:00
22:00
23:00
22:30
0:00
1:00
0:30
1:30
2:30
2:00
3:00
4:00
3:30
4:30
5:30
5:00
6:00
7:00
6:30
10:30
10:00
7:30
11:00
8:00
12:00
11:30
9:00
8:30
12:30
9:30
13:30
13:00
14:00
15:00
14:30
15:30
16:30
16:00
17:00
18:00
17:30
18:30
19:30
19:00
20:00
21:00
20:30
21:30
22:00
23:00
22:30
0:00
1:00
0:30
1:30
2:30
2:00
3:00
4:00
3:30
4:30
5:30
5:00
6:00
10:00
7:00
6:30
10:30
7:30
11:30
11:00
8:30
8:00
12:00
9:00
12:30
9:30
13:30
13:00
14:00
15:00
14:30
15:30
16:30
16:00
17:00
18:00
17:30
18:30
19:30
19:00
20:00
21:00
20:30
21:30
22:30
22:00
23:00
0:30
0:00
1:00
2:00
1:30
2:30
3:00
4:00
3:30
4:30
5:30
5:00
6:00
10:00
7:00
6:30
10:30
7:30
11:30
11:00
8:30
8:00
12:00
9:00
12:00
12:30
9:30
13:30
14:30
14:00
15:00
16:00
15:30
16:30
17:00
18:00
17:30
18:30
19:30
19:00
20:00
21:00
20:30
21:30
22:30
22:00
23:00
0:30
0:00
1:00
2:00
1:30
2:30
3:30
3:00
4:00
5:00
4:30
5:30
6:30
6:00
10:00
7:00
10:30
7:30
11:30
11:00
8:30
8:00
12:00
9:00
13:00
12:30
9:30
13:30
14:30
14:00
15:00
16:00
15:30
16:30
17:30
17:00
18:00
19:00
18:30
19:30
20:30
20:00
21:00
21:30
22:30
22:00
23:00
0:30
0:00
1:00
2:00
1:30
2:30
3:30
3:00
4:00
5:00
4:30
5:30
6:30
6:00
10:00
7:00
11:00
10:30
8:00
7:30
11:30
8:30
12:00
9:30
9:00
13:00
12:30
13:30
14:30
14:00
15:00
16:00
15:30
16:30
17:30
17:00
18:00
19:00
18:30
19:30
20:30
20:00
21:00
22:00
21:30
22:30
23:00
0:00
0:30
1:30
1:00
2:00
2:30
3:30
3:00
4:00
5:00
4:30
5:30
6:30
6:00
10:00
7:00
11:00
10:30
8:00
7:30
11:30
8:30
12:30
12:00
9:30
9:00
13:00
14:00
13:30
14:30
15:30
15:00
16:00
16:30
17:30
17:00
18:00
19:00
18:30
19:30
20:30
20:00
21:00
22:00
21:30
22:30
23:00

01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
Date

Figure 3.15 Hs and wind direction against time (Watson 2001)

N N

D
i
W W
r
e
c
t
i S S
o
n

E E
5 10 15 20 25 30 35
Wind Speed (kts)

1.00-1.50 1.50-2.00 2.00-2.50 2.50-3.00 3.00-3.50 3.50-4.00 4.00-4.50 4.50-5.00

Figure 3.16 Hs (m) against wind speed and direction (Watson 2001)

74
3.3 Study of Waves Arriving at Cape Sorell

Cape Sorell, about midway on the west coast of Tasmania, is the only other exposed
location in southern Australia where the Bureau of Meteorology has a wave rider
buoy in operation. Originally it was not intended to study this data but as the project
progressed, the scope altered and the value in acquiring this data became apparent.
Also is became clear that other buoy data, particularly for Portland, Victoria, was
difficult to obtain. The Cape Sorell wave rider buoy is located at the coordinates
42.12S, 145.03E or approximately 10 km west of Cape Sorell where the depth is 100
metres. The geography of Cape Sorell relative to Cape Du Couedic is shown in
APPENDIX I. It is exposed to powerful swell from a large fetch area illustrated in
Figure 2.4, producing some of the highest wave energy levels in all Australia. The
buoy is the same type as the one located near Cape Du Couedic, also manufactured by
Datawell, but was deployed some years earlier. The analysis in this section is not done
in as much detail as that of Cape Du Couedic, but follows the same format and makes
comparisons at different stages. Given that a longer period of data is available it is
easier to make and confirm long term trends in the analysis.

3.3.1 Data Description

Seven years of data was acquired covering the 1st of January 1998 to the 31st of
December 2004 and it has the same format as the Cape Du Couedic data. As before,
some holes and discrepancies were located in the data and Table 3.3 shows the total
count of recordings each month with months containing discrepancies highlighted.
The first recordings have intervals of 20 minutes but this changes to 30 minutes on the
19th of March 1998 and remains so through to the end of the data set. There are a few
small gaps during February, March and April 1999 and in November 2003 accounting
for the lower count in those months. At 9am on the 12th of October 1999 the
recordings stop and then continue at 3pm on the 10th of November 1999. During 2004
there were some big gaps from the 30th of April to the 8th of May and then on the 30th
of May to the 29th of July. There may be some other small gaps but the data for the
years 2000, 2001, 2002 and 2003 are reasonably complete sets.

75
Table 3.3 Total number of recordings per month 1998-2004
Jan-98 1730 Jan-99 1486 Jan-00 1485 Jan-01 1457 Jan-02 1484 Jan-03 1484 Jan-04 1474
Feb-98 1550 Feb-99 1129 Feb-00 1388 Feb-01 1309 Feb-02 1329 Feb-03 1338 Feb-04 1383
Mar-98 1546 Mar-99 1292 Mar-00 1466 Mar-01 1469 Mar-02 1453 Mar-03 1469 Mar-04 1474
Apr-98 1428 Apr-99 1050 Apr-00 1424 Apr-01 1420 Apr-02 1422 Apr-03 1416 Apr-04 1391
May-98 1479 May-99 1471 May-00 1400 May-01 1473 May-02 1464 May-03 1477 May-04 1050
Jun-98 1426 Jun-99 1418 Jun-00 1420 Jun-01 1401 Jun-02 1388 Jun-03 1425 Jun-04 0
Jul-98 1483 Jul-99 1483 Jul-00 1480 Jul-01 1352 Jul-02 1457 Jul-03 1479 Jul-04 116
Aug-98 1444 Aug-99 1470 Aug-00 1478 Aug-01 1462 Aug-02 1470 Aug-03 1475 Aug-04 1207
Sep-98 1434 Sep-99 1411 Sep-00 1396 Sep-01 1429 Sep-02 1388 Sep-03 1328 Sep-04 1436
Oct-98 1481 Oct-99 383 Oct-00 1474 Oct-01 1472 Oct-02 1478 Oct-03 1486 Oct-04 1481
Nov-98 1436 Nov-99 978 Nov-00 1433 Nov-01 1434 Nov-02 1427 Nov-03 935 Nov-04 1433
Dec-98 1476 Dec-99 1421 Dec-00 1439 Dec-01 1484 Dec-02 1474 Dec-03 1478 Dec-04 1484

3.3.2 Statistical Analysis and Seasonal Climate Trends

Yearly means for significant wave height and zero crossing period were taken for all
of the data and are displayed in 3.4 along with the averages over the 7 year period.
The values for 2004 may be slightly underestimated due to missing data during winter
months. The year to year variations of Hs appear to be about 30 cm or about a 20%
swing in available power and in a cycle between 1998 and 2002. As anticipated, the
wave energy levels at Cape Sorell are significantly higher than those at Cape Du
Couedic. Overall the average Hs is 30 cm higher and Tz is 0.2 seconds longer at Cape
Sorell, with an estimated 9.7kW/m or 30% more available power. As with Table 3.2,
the averaged values are simply the mean of the other values, not the overall mean.
Also note that calculating available power from the mean Hs and Tz values for each
year results in a value significantly less than the actual mean available power listed
for that year (calculated at each recording). More detailed comparison follows.

Table 3.4 Summary of Mean Values at Cape Sorell


Mean Value 1998 1999 2000 2001 2002 2003 2004 Averaged
Hs (metres) 3.04 2.85 3.07 2.79 3.05 2.97 2.78 2.94
Tz (seconds) 8.01 7.90 8.02 7.88 7.84 7.92 7.81 7.91

Available Power (kW/m) 44.80 38.91 47.26 38.42 46.61 43.65 36.98 42.38

76
The largest recorded wave occurred at 11:30am on the 10th of June 2002 with an
amplitude Hmax= 18.31 metres and period Tmax of 13.78 seconds. Using equation (2.1)
this wave would have had a power flux equal to 4,422kW/m. This wave occurred 10
days before the largest wave was recorded in the Cape Du Couedic data. It is likely
that the long fetch winds were very strong in June 2002 and these powerful waves
were generated by separate local storms.

Significant Wave Height Cape Sorell Zero Crossing Period Cape Sorell
Cape Du Couedic Cape Du Couedic
4 9
3.75 8.75
3.5 8.5
3.25 8.25

seconds
3 8
metres

2.75 7.75
2.5 7.5
2.25 7.25
2 7
1.75 6.75
1.5 6.5
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Available Power Flux Cape Sorell Cape Du Couedic


70
65
60
55
50
45
40
kW/m

35
30
25
20
15
10
5
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Figure 3.17 Combined monthly mean plots and comparison

All the mean values for each month of the 7 years were combined and plotted in
Figure 3.17. The Cape Du Couedic curves from Figure 3.3 are also included for the
sake of easy comparison, even though they only represent 3 years of data. The pairs of
curves follow the same annual trends with Hs and available power being higher at
Cape Sorell overall and Tz periods being more equal. The peak of wave energy at both
locations is pronounced during September and levels appear to drop between June and
September. Winter in southern Australia is officially the months of June, July and
August. Although after winter, September is typically a period when there are
powerful and consistent winds over the Southern Ocean fetch area resulting in more
powerful westerly swell. The Cape Sorell data shows that the wave energy between
January and June is a more gradual climb, whereas at Cape Du Couedic wave energy

77
levels remain a lot calmer until winter. As mentioned in section 3.2.2, this plot can not
be used as an accurate prediction tool due to the variability month to month and year
to year, best illustrated in Figure 3.18. The vertical gridlines are placed at the start of
every year to help visualize annual cycles. For reference and easy comparison the
Cape Du Couedic curve is also included. The correlations between the two sets of
data are quite close for monthly means, with Tz being fairly even and Hs being usually
higher at Cape Sorell. Each year the number of peaks for monthly mean Hs varies and
not every year do they exceed 4 metres. It is not uncommon to have regular periods in
winter months with relatively calm wave energy, which might not be expected at
given Cape Sorell is one of Australia’s wildest coasts. A powerful storm during any
month of the year can greatly increase the values of mean Hs. A few histograms were
generated for some years of the Cape Sorell data that showed a very similar
distribution to those for Cape Du Couedic in 2003 in Figures 3.6, 3.7 and 3.8 except
mean values are slightly higher.

6
Cape Sorell Hs (metres) Cape Du Couedic Hs
Cape Sorell Tz (seconds) Cape Du Couedic Tz
5

0
Jan-98 Jan-99 Jan-00 Jan-01 Jan-02 Jan-03 Jan-04

Figure 3.18 Monthly means over the 7 years of data

78
3.4 Survey of Australian Coastline Bathymetry

A very important factor for determining the suitability of a wave farm location is the
distance from the shoreline to the 50m depth contour. Ideally the minimum depth
needed for the Pelamis mooring as between the 50-100m contours only a small
proportion of swell wave energy is lost. The distance from the shoreline to the 50m
contour dictates the minimum length of submarine electrical cable needed. This is
important because if the cable is more than about 10km long, its cost alone is likely to
exceed the total cost of every other aspect of a pilot wave farm. From the viewpoint of
wave energy resource, not only is the location of this contour important but also its
exposure to long fetch swell and the obstructing bathymetry preceding the 50m
contour, such as islands and reefs. With these limitations in mind, a survey was
undertaken for the entire coast of southern Australia.

To begin with it is worth looking at the ocean contours surrounding Australia at large
as shown in the image in APPENDIX H. This image gives an indication of the deep
ocean contours leading into the Australian plate. There is minimal disruption of wave
energy over the Southern Ocean fetch area (see Figure 2.4) between Australia and
Antarctica in the form of islands or other projections.

As with the case of sourcing wave data, locating bathymetric information for specific
sites can be a time consuming process, not without expense. Several Hydrographic
charts and bathymetric maps where scanned over in process of this survey but they
cannot all be reproduced here due to copyright issues. Hydrographic charts for the
entire Australian coastline are available from the Australian Hydrographic Service.
They are plotted with the location of depth soundings with units typically in fathoms
(6ft). The distance of the sounding from the shoreline can be measured using the scale
of the map. Given that a number of specific and widely separated locations are
mentioned, there will be no bathymetric maps included for reference in this summary
of the bathymetric survey. Sections of bathymetric maps have been used in the Case
Studies in Chapter 4. Some of the areas discussed can be located on the image in
APPENDIX I. Initially the focus was on Kangaroo Island but then extended to the
rest of SA and then VIC, TAS and WA. The results of this survey will be summarised
by tracing the coastline to the east of Kangaroo Island and then to the west.

79
The following is a list of hydrographic charts in the order they are used in this
summary.

• AUS346 Kangaroo Island (South Coast)


• AUS347 Backstairs Passage to Cape Martin (Coorong & Southeast SA)
• AUS349 Cape Nelson to Cape Schanck (Portland and Southwest VIC)
• AUS789 King Island
• AUS353 Hunter Island to Low Rocky Point (Tasmania West Coast)
• AUS342 Streaky Bay to Whidbey Isles (Eyre Peninsula West Coast)
• AUS335 Cape Naturaliste to Point D’Entrecastauex (Southwest WA)

Kangaroo Island would create a fetch limiting shadow for the lower section of the
Fleurieu Peninsula from south westerly swell. Also this area, Backstairs Passage, is
mostly shallower than 50m. The details of Kangaroo Island’s bathymetry are covered
in section 4.1.2. To the immediate east of Kangaroo island is the Coorong region
extending from Goolwa to Kingston. This entire region has a very shallow drop in the
sea floor with the 50m contour typically more than 40kms out. As a result the wave
energy levels arriving at these shores are comparatively low. The 50m contour does
move closer to the coast after Kingston but the nearest point is just over 10kms
adjacent to the Canunda Wind Farm (near Mt Gambier). After there the contour
moves out further into Discovery Bay and into Victoria. Cape Duquesne near Portland
is the next point east of Cape Du Couedic where the 50m contour moves within 3kms
of the shore. Details of bathymetry in the Portland region are shown in section 4.2.1.
East of Portland the 50m contour moves beyond 10kms from the shore up to and
beyond the entrance of Port Philip Bay.

King Island is an island of Tasmania situated to its Northwest in the Bass Straight.
The focus is on the west coast only where there is the best exposure to long swell
wave energy. About halfway down the coast is the town Currie, not far south from it
is a 50m sounding 4.5 km out. Further south of Currie the 50m contour moves closer,
at the South western tip it is as near as 1-2kms. To the north of Currie the connection
would not be so good because the 50m contour moves further out and in places there
are small islands obstructing the swell.

80
The west coast of Tasmania has a series of capes and headlands with steep bathymetry
interspaced by shallower bays. Some of the more prominent and exposed locations are
mentioned here. The chart for this region (AUS353) has fairly sparse soundings and
estimations were made for the position of the 50m contour because it is not indicated.
Starting at the Northwestern tip, Gape Grim is about 6 km from the 50m contour.
Moving south to West Point nearer to the town of Marrawah there is a 72m sounding
4.5 km out. Further south to Sandy Cape near the mouth of the Pedder River the 50m
contour is about 4.5km out, although it is sheltered by a rise indicated by a 30m
sounding 6km out. Cape Sorell not far from the town of Strahan has two 70m
soundings radiating just over 6 km out and the 50m contour is estimated at about 5 km
out. No more bathymetric information was obtained South of Cape Sorell. The east
coast of Tasmania and the coasts on remaining parts of the eastern states receive
considerably less wave energy due to limited fetch, so the survey looked no further
east.

Now looking to the west of Kangaroo Island, the southwestern coast of the Eyre
Peninsula would receive a fairly similar wave climate to Cape Du Couedic. Cape
Carnot at the southern tip is 2kms from a 50m contour, although there would be a
small amount of sheltering by surrounding islands and protrusions. Whidbey Point on
the Coffin Bay Peninsula has 3kms to the 50m contour although it would have minor
sheltering by some small protrusions and islands, in particular Greenly Island 30km to
the west. Further north and adjacent to where Lake Hamilton meets the coast, the 50m
contour is 4kms out with relatively few obstructions. Between Tungketta Reef and
Elliston the 50m contours moves to 3kms to shore at some locations but waves would
be significantly obstructed by a number of protrusions out to Flinders Island and the
Investigator Group. North of Elliston is Anxious Bay, where the 50m contour moves
between 3-5kms from the shore with minimal wave obstruction. Just above Anxious
Bay there are a number of locations where the 50m contour moves close to 2km but
the ones with the least westerly obstruction are Cape Radstock and Point Westall,
which is 20km from Streaky Bay. After Point Westall the 50m contour moves far out.

Westerly fetch becomes limited further up the Eyre Peninsula and wave energy levels
reduce. Between the head of the Great Australian Bight and the south west of WA the
westerly fetch is small or non existent leading to low wave energy levels. Figure 3.1

81
and the other ERA-40 plots in APPENDIX E support this theory and this is why the
bathymetry of this section of coastline was not explored. The west facing region of
the southwest corner of WA would be exposed to very long westerly fetch, perhaps
with similar long swell energy to Tasmania. The region between Cape Naturaliste to
Point D’Entrecastauex was investigated (chart AUS335). It appears the entire stretch
of this coastline is skirted by large reefs, the Naturaliste and South West Reefs
respectively. Although Cape Naturaliste has a 50m contour 3km from shore and Point
D’Entrecastauex 6km from shore, there is a substantial amount of shallower water
further out, up to 30m. In both cases the shortest line to the ‘outer’ 50m depth contour
is 24km and these represent the closest connection points in this region. Waves would
be stripped of a large amount of energy before reaching these ‘inner’ 50m depth
contours. Only site specific data could be used to determine if these connections are
of good potential.

3.5 Measure Correlate Predict Model for Waves

For a site specific analysis of wave energy it is important to obtain accurate and
reliable wave data. Wave rider buoys are still the most accurate portrayal of wave
conditions at a particular site and provide the most useful information. However,
wave rider buoys can be few and far between, as is the case in southern Australia. The
alternative is to use modeled data or data acquired or supplemented by satellites. If
there are two or more sets of available data from wave rider buoys near or
surrounding a desired site a process known as measure correlate predict (MCP) can be
used to model the site. This process is well established in the wind energy industry
where data must be modeled over large sites with great detail. For a wind MCP
correlations are good (R=0.95) for data separated by 100kms. In conjunction with this
project, Ocean Prospect has been developing a MCP model applicable to wave
energy. Due to the lack of available data it could not be made use of for this project,
but may find uses in future applications on Australian coasts. Cape Sorell and Cape
Du Couedic are too far apart for an accurate and realistic MCP model with good
correlation.

82
3.6 Conclusions

The wave energy arriving at southern Australian coasts is among the highest in the
world, exceeded or rivaled only by the coasts of South America (Southern Chile),
Northern Europe (Great Britain), New Zealand (West Coast of South Island) and
South Africa, see Figure 3.1 and APPENDIX E. The general trend for southern
Australia is for wave energy levels to increase further south with the wildest waves
arriving at south west Tasmania. The analysis of waves arriving at Cape Du Coeudic
and Cape Sorell provide an indication of the wave energy levels and the wave
climatology of southern Australia.

Wave energy levels are dictated by seasonal cycles and the variable wind speeds and
directions in the fetch area of the Southern Ocean. Daily mean wave energy levels
vary by over 16kW/m on average and can fluctuate between 20 and 110kW/m several
times per month during winter. From the analysis in this chapter, it can be estimated
that the long term mean wave energy, uninterrupted by bathymetry, arriving mid way
along the coasts of South Australia are 32 kilowatts per meter (mean Hs=2.6m,
Tz=7.7s). Further south at Cape Sorell the wave energy levels are among the highest
in Australia with a long term mean power flux of about 42 kilowatts per metre. Mean
long swell direction is south west to westerly at South Australia and more directly
westerly down at Tasmania. Seasonal and monthly trends correlate quite closely
between Cape Du Couedic and Cape Sorell. An important observation is made in
Figure 3.14 that there can be strong wave energy when winds are calm or dead. This
is particularly the case when winds drop suddenly.

The bathymetric survey shows that there are a limited number of sites in southern
Australia which are close enough to the 50m sea depth contour to make an electrical
connection viable. Combining the bathymetric survey with the indicated wave energy
levels, the best available connection points for a Pelamis wave farm can be
highlighted. Portland appears to have the best conditions for a wave farm connection
on mainland Australia with high exposure to wave energy and steep local bathymetry.
Portland is situated roughly halfway between Cape Sorell and Cape Du Couedic and
would likely receive wave energy levels halfway between those two locations. The
case study for a Portland Pelamis wave farm is presented in section 4.2.

83
Tasmania’s west coast has a number of potential connection sites, Cape Sorell being
the most attractive due to its proximity to demand at Strahan. King Island has good
potential on the west coast near Currie and further south. The feasibility of wave
energy projects in Tasmania is discussed in section 4.3.1. The feasibility of a wave
farm at Cape Naturaliste or Point D’Entrecastauex in south west WA is questionable
due to reefs and other obstructing bathymetry leading to uncertain wave energy levels
nearer the shore. The west coast of the Eyre Peninsula of South Australia has a
number of sites which could lead to a wave energy project, some with less westerly
obstruction than others. Potential projects for the Eyre Peninsula are discussed in
section 4.3.2.

84
4 Case Studies for Potential Wave Farms
Initially the focus of this project was on the feasibility of developing a wave farm for
Kangaroo Island, but it soon became apparent that the case for Kangaroo Island was
less promising than originally expected. The scope has broadened to include potential
wave farms over the entire southern Australian coastline. This chapter deals mostly
with the feasibility of mains connection for the locations highlighted in the conclusion
of the previous chapter.

4.1 Kangaroo Island Wave Farm Study

From the analysis in Chapter 3 it can be seen that there is an abundant source of wave
energy at Cape Du Couedic, Kangaroo Island. This case study investigates the
feasibility of a real wave farm for Kangaroo Island and how it could be connected to
the grid. Figure 4.1 shows the distribution of Kangaroo Island’s electricity grid
network.

Figure 4.1 Kangaroo Island’s Electricity Grid Network (source ETSA Utilities)

85
4.1.1 Summary of KI Energy Review

Kangaroo Island, located off SA’s southern Fleurieu Peninsula (refer to Figure 4.1)
has approximately 4500 permanent residents (≈0.3% of SA’s population) and is
geographically isolated from the Australian mainland. The island has a history of
problematic issues with the reliability and the capacity of its energy supply. The
Kangaroo Island Development Board (KIDB) based in Kingscote, is currently in the
process of reviewing its energy needs. In June 2003, a publicly available report was
released called the ‘Kangaroo Island Energy Review - Stage 1 (SWWES 2003)’. The
report was prepared for and funded by the KIDB and the State Energy Research
Advisory Committee (SENRAC) with assistance from ETSA Utilities and ESIPC.
The following is a summary of this report (SWWES 2003) relevant to this project.

Kangaroo Island’s main electrical load is concentrated around the Kingscote area. The
island’s base load electricity demand was 2.5MVA and the current peak load was
estimated to be 6.3-6.8MVA for 2005, compared to SA’s total peak of over 3000MW.
Kangaroo Island’s primary source of electrical power is from the mainland via a
submarine power cable connected between Cape Jervis and Penneshaw (see Figure
4.1), that had a replacement value estimated at $4-6 million (SWWES 2003). ETSA
Utilities quoted in a recent meeting that this figure is more like $15-20 million, which
includes installation and connection costs. The maximum load that can be distributed
to the island through the existing cable is 5.4MVA, which is restricting the island’s
peak usage and economic growth. The SWWES report contains details of the island’s
electricity grid and infrastructure. According to ETSA Utilities, the island is currently
unable to export any electrical power to the mainland.

The island has a diesel back-up generator in place with a capacity of 2.4MW and if
the submarine cable were to fail, KI’s electrical power would be solely reliant on this
generator. The threat posed by the unexpected complete failure of the submarine cable
is real. ETSA Utilities has approved the proposal for a new 6.5MW remotely
controlled backup diesel generator at the cost of $10 million (Kirke 2004) to replace
the old one. This is an expensive insurance policy that would be funded by the state,
but would provide KI with a backup capacity of 6.5MW in the event of cable failure.

86
The KIDB sees the perception of the island being “clean green” as a major asset and
is eager to supply the island with renewable energy. The report (SWWES 2003) states
that KI is rich with a range of renewable energy sources, including ocean waves. The
stage 2 report, which was to be commissioned, was to explore these options in greater
detail. A recent report by Kirke (2004) of UniSA’s SEC investigates some energy
supply issues and renewable energy solutions for the island.

4.1.2 Grid Connection and Location Bathymetry

According to ETSA Utilities there are 4 substations on the island, which are named in
Figure 4.1. The best way to insert externally generated electrical power into a network
is through a substation. At the commencement of this project it was envisaged that a
connection could be made for a wave farm at American River. ETSA Utilities
confirmed that the American River substation is able to accept external electricity
input. This led to an exploration of the bathymetry off the coasts near this substation
as shown in Figure 4.2.

Figure 4.2 Coastal bathymetry of the region south of American River


(Source: Chart AUS346, copyright of Australian Hydrographic Service 1992)

87
The depth soundings shown on this extract of a hydrographic chart are in fathoms
(6ft). The 50 meter contour (27.34 fathoms) has been sketched in place indicating
where the minimum depths for Pelamis mooring occur. The closest 50 meter depth is
about 20km from the nearest point of the coast to the American River substation, near
Reynolds Point. A submarine cable laid over this region would be of similar length
and expense to the one connecting Cape Jervis to Penneshaw. In addition to this there
would be significant sheltering by Cape Gantheaume from westerly swell and
obstructing bathymetry from regions further west. Wave energy levels at this region
would be very difficult to model accurately given only bathymetric data and
waverider buoy data from Cape Du Couedic. However it can be almost certain there
would be a lot less wave energy arriving at this location compared to Cape Du
Couedic from sheltering and attenuation in the bay. All of this indicates that the
feasibility of a wave farm connection to American River is not attractive.

Ideally a connection could be made at Cape Du Couedic where the highest wave
energy levels are but unfortunately the grid is not set-up for it. For reference, a
connection from Cape Du Couedic to the 50 metre contour is about 3kms. There is a
spread of 19kV SWER (single wire earth return) lines over to the west side of the
island and even one running right down to Cape Du Couedic. However a wave farm
connection can not be made through a 19kV cable as they are single phase. Upgrading
the power lines and poles to the required 33kV cable to make a connection from
Macgillivray substation to Cape Du Couedic (see Figure 4.1) would certainly be a
prohibitively expensive process. There would also likely be disputes over who owns
sections of the grid and the land it occupies. Cape Du Couedic and a large portion of
the west of the island is Flinders Chase National Park. There may also be issues with
building grid down to make a connection at Cape Gantheaume, which is a
conservation park. Cape Hart on the east of the island (see Figure 4.2) has a 50m
depth contour about 8kms from shore but it would be very sheltered from westerly
swell and there is no existing grid to connect it to American River.

88
4.1.3 Pelamis Performance Model at Cape Du Couedic

This section looks at the output of a single Pelamis unit placed at the location of the
Cape Du Couedic buoy over the 3 year period. The Pelamis is not going to be located
at the Cape Du Couedic buoy for reasons described in the previous section, so only a
good approximation is needed. Cape Du Coudic is a representation of the kind of
wave climate arriving to a fairly large section of the southern Australian coastline, so
it would be useful to determine likely capacity factors for the Pelamis at this location.
Instead of multiplying every data value through the performance matrix, a simpler and
quicker method was devised. Firstly a plot was produced for a number of different Tz
values in the performance matrix in Figure 2.15 and is shown as Figure 4.3 below.

800

700
3.5s
4.5s
600
5.5s
6.5s
500
Power (kW)

7.5s
400 8.5s
9.5s
300 10.5s

200

100

0
0 1 2 3 4 5 6 7 8 9 10
Significant Wave Height (m)

Figure 4.3 Pelamis Performance Curves for various Tz values

Even though a similar plot for available power flux shows parabolic curves, the power
curves from the Pelamis are mostly linear. Looking at Tz histograms in Figures 3.6 &
3.9, the curve for Tz = 3.5 seconds can be ignored due to such conditions being so
extremely unlikely (Figures 3.6 & 3.9). Zero crossing periods of 10.5 and larger are
also quite rare. The monthly means from all of the data fit within the ranges
3.8>Hs>1.8 metres and 8.5>Tz>6.5 seconds. A closer look at the curves for these
ranges is provided in Figure 4.4.

89
6.5s 7.5s 8.5s
600
550
500
450

Power (kW)
400
350
300
250
200
150
100
1.8 2 2.2 2.4 2.6 2.8 3 3.2 3.4 3.6 3.8
Significant Wave Height (m)

Figure 4.4 Near Linear Property of Tz curves

Through observation these curves can be closely approximated by fitting linear


equations of the form y = aχ + b. The chosen equations are shown as follows:
Power = 190 H s − 172 8.0>Tz>6.5 (4.1)

Power = 175 H s − 190 8.5>Tz>8.0 (4.2)


Using these relationships, Pelamis power output values were calculated for each of
the monthly mean values for the whole data set and then graphed in Figure 4.5. Again
the value for May 2002 is missing. Undoubtedly this approach has its shortcomings as
it is based on monthly means only and doesn’t consider the other ‘less steep’ Tz
curves. It should however be reasonably accurate for the purposes of trends and
capacity factors. Figure 4.6 shows the same data over one year for easier comparisons.
Mean power outputs were taken from the other 3 years and are also plotted in Figure
4.6 with a 5th degree polynomial trend line.

500
450
400
Power Output (kW)

350
300
250
200
150
100
50
0
Nov-00

Jun-01
Jul-01

Nov-01

Jun-02
Jul-02

Nov-02

Jun-03
Jul-03

Nov-03

Jun-04
Aug-00

Jan-01

May-01

Aug-01

Jan-02

May-02

Aug-02

Jan-03

May-03

Aug-03

Jan-04

May-04
Sep-00
Oct-00

Dec-00

Feb-01
Mar-01
Apr-01

Sep-01
Oct-01

Dec-01

Feb-02
Mar-02
Apr-02

Sep-02
Oct-02
Oct-02

Dec-02

Feb-03
Mar-03
Apr-03

Sep-03
Oct-03

Dec-03

Feb-04
Mar-04
Apr-04

Figure 4.5 Mean monthly Pelamis output over the whole 3½ years

90
500
2001
450
2002
400 2003
Pelamis Power Output (kW)

Means
350

300

250

200

150

100

50

0
jan feb mar apr may jun jul aug sep oct nov dec

Figure 4.6 Annual Comparison of Mean Pelamis Outputs

More data and a more accurate analysis would be needed for this graph to be used as
an accurate long term prediction tool. It does indicate that summer mean power
outputs are in the order of 250kW and in winter the mean values are from 350 to 400
kW, but peaking in September. This graph has a trend bearing resemblance to the
power flux curve in Figure 3.3. The mean capacity factors are displayed in Table 4.1.

Table 4.1 Summary of Mean Capacity Factors for Pelamis at CDC


Year Annual Mean Power Capacity Factor
2001 294.9 kW 39.3%
2002 325.7 kW 43.3%
2003 303.4 kW 40.4%
All Data 303.6 kW 40.4%

As stated earlier, the Pelamis will not be located at the CDC buoy so this is only an
indication of what the Capacity Factor might be like for the wave energy levels at this
location of the South Australian coast.

Firm Capacity is a term often used in the wind industry to state a Capacity Factor
which can be relied upon with 95% confidence. It can only be calculated using long
term data and taking into account when power is not being generated due to

91
insufficient available energy or cut out due to excess energy (or in the case of the
Pelamis – limited to 750kW). There is not enough years’ worth of data available and
too much variation year to year to make a good estimate of Capacity Factor with 95%
confidence. It is worth, however, investigating how much time the Pelamis spends
idle or at the 750kW capacity that has not been taken into account in the above
calculations. Using the performance matrix Figure 2.15, an assumption can be made
that the Pelamis is always idle when Hs<0.5 metres. The histograms in Figure 3.6 and
3.9 and the scatter diagram Figure 3.12 all show that this condition is an extremely
unlikely occurrence at Cape Du Couedic. There are conditions when above Hs=0.5
metres the Pelamis will be idle if Tz is at or below 4.5 seconds, perhaps 1% of the
time, or above 8.5 seconds, occurring almost never according to the scatter diagram.
Likewise looking at the region above the line plotted on the scatter diagram, there is
not one point located at a condition where the Pelamis generates 750 or more
kilowatts according to the power matrix. The year 2002 was wilder than 2003 (used
for the scatter diagram), but it still shows that the likelihood of the Pelamis generating
zero output or more than 750kW is very rare, perhaps up to one percent of the time.
This adds strength to the Capacity Factor calculations made above.

4.1.4 Recommendations

The south west coast of Kangaroo Island is best exposed to long swell wave energy,
sufficient to allow a single Pelamis to deliver a mean annual power output in the order
of 300kW or total electrical energy production of 2.63GWh. All of the other coasts on
the island receive significantly less wave energy by being sheltered from westerly
swell. Unfortunately the island’s electricity grid is not equipped to accept large
insertions of electrical generation on the west side of the island. The substations and
demand are located on the east of the island where a combination of bathymetry and
sheltering make a poor site for a Pelamis wave farm. This leads to the conclusion that
the overall economics of a pilot wave farm of 4 Pelamis are not commercially
attractive for Kangaroo Island due to very high connection costs.

The KI Energy Supply Options report by Kirke (2004) explores some alternative
renewable energy supply options for Kangaroo Island. Wind Energy is prevalent on
the island but only 8% Firm Capacity is estimated. The report indicates there is a
potential for pumped hydro storage at Middle River Dam, and tidal current energy in

92
Backstairs Passage. Perhaps the most promising project in the short term is to use
locally grown biomass or biowaste to produce energy with zero net CO2 in mind.
Biodiesel could be produced from canola to power the islands diesel generators.
.
4.2 Portland Wave Farm Study

This section investigates the feasibility of a Pelamis wave farm project situated near
Portland, Victoria. Portland has an abundant source of renewable energy and is
quickly becoming a centre for new projects. The Glenelg Shire Council, which is the
Local Government Authority for Portland, has produced a letter to the government
reporting the enthusiasm for Portland’s renewable energy visions and projects (Purton
2003). The letter highlights that Portland is a very attractive site for wave energy due
to strong swell waves close to high capacity electricity grid. Portland is already the
site for a pilot wave energy project by Ocean Power Technology (OPT 2005). The
IPS buoy is a point absorbing wave energy converter with each unit rated at 20kW.
According to the letter by Purton (2003) this pilot plant was due to be operational
during 2003. Local community and government support were very supportive of this
pilot project. Wind energy has become the focus at Portland with the Portland Wind
Energy Project recently being approved for development by Pacific Hydro. The 195
MW wind farm will not only be the largest in Australia, but the entire southern
hemisphere. This may be a source for competition with the Pelamis, but the advantage
here is that wave energy is usually still strong with sudden drops in wind energy (see
Figure 3.14). Danish turbine manufacturer Vestas has begun constructing a blade
manufacturing facility at Portland that looks to provide a boost to the local economy.

4.2.1 Location Bathymetry and Grid Connection

As indicated in the conclusion of Chapter 3, Portland is the best site for wave energy
on the mainland. It is highly exposed to westerly and southerly swell and has steep
bathymetry leading into the shore. Figure 4.7 show the bathymetry surrounding
Portland. The 50 metre depth contour is about 1.5kms from shore at both Cape
Duquesne and Cape Nelson, and about 2kms out from Cape Bridgewater. Each of
these locations would be very good for a wave farm, perhaps the most exposed site
lies just off Cape Duquesne.

93
Figure 4.7 Coastal bathymetry surrounding Portland with depths in metres
(Source: Sheet SJ 54-11, copyright of NATMAP 1987)

Figure 4.8 Cape Duquesne Location (Plummer et al. 2003) – modified

94
Another aspect to consider is that this region is of importance for marine wildlife
conservation. Parks Victoria has published a report (Plummer et al. 2003) with details
of the Discovery Bay Marine National Park shown in Figure 4.8. Laying a submarine
cable within this section of seabed off the shore of Cape Duquesne is likely to be
prohibited. Figure 4.9 also shows depth contours and a target has been placed at the
site which would be ideal for a wave farm 1.5 km off Cape Duquesne outside the
marine park. Cape Nelson State Park in Figure 4.7 is also managed by Parks Victoria,
but given that wind turbines have been approved on this location it is unlikely that this
will cause any problems with connecting a wave farm. In any case, Parks Victoria will
need to be notified and consulted for planning a wave farm for Portland.

An initial enquiry has been made to Powercor (equivalent of ETSA Utilities but for
Victoria) regarding a wave farm connection off Portland. Figure 4.9 is an excerpt
from a drawing supplied by Powercor, it shows the electricity grid network
surrounding Portland. Powercor have a 66/22kV zone substation close to Portland
township marked as a square named “PLD” in Figure 4.9. According to Powercor, the
stability and magnitude of the generation would determine the connection in this are
and only a detailed technical study would provide a final decision. Powercor has
expressed positive interest in assisting with the proposed Pelamis wave farm.

Figure 4.9 Electricity Grid surrounding Portland (Source: Powercor)

95
4.2.2 Resource and Performance Prediction

Quite some time was spent searching for and pursuing wave data for Portland.
Research indicated that CSIRO may have or once had wave rider buoys located at
Portland, Mt Gambier and Streaky Bay. Information about these buoys and their data
was very difficult to obtain. This led to an exploration of other options for obtaining
wave data for the Portland site. One possibility is to purchase modeled wave data
from the NOAA Wave Watch III (WW3). It would provide over 8 years of data
including wave heights, periods and directions along with modeled wind speed and
direction available on a 1.00 x 1.25 degree grid spacing. Good accuracy can be found
with the NOAA WW3 in depths of 50 to 100 metres and is routinely verified by a
number of buoys and weather ships worldwide. Portland also a practical location for
modelled data due to there being no obstructing bathymetry (reefs, islands etc.).
However, prices have been quoted for this data and Wind Prospect is not prepared to
make the purchase at this current stage.

This difficulty in obtaining data from Portland was the main motivation behind
obtaining the Cape Sorell data from the Bureau of Meteorology in order to draw some
kind of reference to which Portland’s wave energy levels could be estimated. Cape Du
Couedic and Cape Sorell are almost exactly 1000 kilometres apart and almost half
way between lies Portland. The geography of these locations and distances is best
illustrated in APPENDIX I. Given that the monthly means for Cape Du Couedic and
Cape Sorell correlate reasonably well, it can be postulated that the wave energy levels
arriving at Portland would be roughly half way between that which arrives at these
two locations. It was decided to apply the basic Pelamis performance model derived
in section 4.1.3 to the Cape Sorell data. The combined monthly means of the 7 years
of Cape Sorell data are compared with the 3½ years of Cape Du Couedic data over
one year and displayed in Figure 4.10. This graph suggests that on average there are
higher wave energy levels at Cape Du Couedic during the month of August, although
this may not be the case. The accuracy limitations of this model are explained in
section 4.1.3 and may be compounded by having only 3 ½ years of data available for
Cape Du Couedic.

96
Pelamis Power (kW) Cape Sorrel Cape Du Couedic
500
450
400
350
300
250
200
150
100
50
0
jan feb mar apr may jun jul aug sep oct nov dec

Figure 4.10 Comparison of Pelamis performance from combined monthly means

Using all of the monthly means from 7 years of data, the average estimated Pelamis
output modeled at the Cape Sorell buoy was calculated at 357.2 kW (Capacity Factor
= 47.6%). This was done by the same means as the Cape Du Couedic estimation,
which was 303.6 kW (Capacity Factor = 40.4%) or 15% less power than Cape Sorell.
The performance of the Pelamis at Cape Sorell may be underestimated given that
close on 30% more wave energy arrives there on average than Cape Du Couedic (a
more precise calculation). Only a more accurate analysis undertaken using a higher
resolution performance matrix, with each data recording multiplied through the
matrix, could determine a more accurate performance output. However, using the two
values now calculated, it is estimated that a Pelamis at Portland would have an annual
mean output in the order of 330kW (Capacity Factor = 44%) or nearly 2.9GWh of
electrical energy.

4.2.3 Recommendations

Portland appears to be the best site for a Pelamis wave farm in Australia and it is
recommended that OPD be notified of this. If a detailed investigation is to be made
into this location by Wind Prospect or OPD, it is recommended that the NOAA WW3
data is purchased and that the Pelamis output be accurately modelled for the site,
Cape Duquesne in particular. This way Powercor can better investigate the feasibility
of inserting wave power to the Portland substation. Parks Victoria will need to be
consulted early on to establish the environmental impact assessment of the sites.

97
4.3 Other Potential Pelamis Projects

If the distance needed for a submarine cable was not such an important limitation for
a potential Pelamis project, the feasibility of other locations would improve. In South
Australia a connection of about 30km could be made at American River, Kangaroo
Island to access good wave energy levels. Also a connection of about 15kms could be
made adjacent to the Canunda Wind Farm, near Millicent, where wave energy levels
would be similar to those arriving at Portland. In Western Australia, the wave energy
levels would also be very high beyond the reefs at Cape Naturaliste over 24 km from
shore. However the connection distance is an important consideration and so is the
need for adequate electricity grid nearby with suitably high energy demand. The
following sections investigate the feasibility of projects regarding nearby demand and
infrastructure that have already been identified as good locations by the bathymetric
survey.

4.3.1 Cape Sorell and other Tasmanian locations

Cape Sorell receives some of the highest coastal wave energy in all of Australia due
to very long fetch illustrated in Figure 2.4. Section 4.2.2 concluded that placing a
Pelamis at the Cape Sorell wave rider buoy would result in a capacity factor of nearly
50%. The nearest demand to Cape Sorell is at the town of Strahan, on the other side of
Macquarie Harbour. A satellite image of this region is provided in Figure 4.11, where
the target symbol shows where a wave farm could be located (the buoy is situated 10
km out). Although this has not been verified, it is very unlikely there is electricity grid
of multi-megawatt capacity in place to distribute power from the most north west
point of Cape Sorell, across Maquarie Harbour and to the town of Strahan. This
infrastructure would need to be built if there were to be any insertion of wave energy.

This region around Strahan, nearby the towns Zeehan and Queenstown would be
where the highest electricity demand is on the western side of Tasmania. The
population of this region is quite small at about 5,000 and there is only a small mining
industry, meaning the energy demand would also be small. Tasmania also has very
large amounts of hydroelectricity, about 40% of primary energy (RPDC 2005).
Energy produced from a Pelamis would need to compete with low cost renewable
hydroelectricity.

98
Most of the other locations mentioned further up the coast from Cape Sorell in the
bathymetric survey would be too far from any significant electricity demand to make
a wave farm feasible. An exception could be at West Point on the north west to supply
energy to the small town of Marrawah. At this location there is competition with the 3
stage Woolnorth wind farms developed by Hydro Tasmania with a generating
capacity of 65 MW already operating (AusWEA 2005). As highlighted in the
bathymetric survey there is potential for a wave farm just south of Currie on King
Island depending on grid infrastructure. Here also there is competition with wind
farms by Hydro Tasmania and the demand is small. The Huxley Hill wind farm has a
capacity of 2.5 MW and has met 20% of the island’s energy demand (AGO 2003).
There are plans to use “large flow” Vanadium Redox batteries to greatly increase the
island’s wind energy usage.

Figure 4.11 Satellite Image of Cape Sorell (source: Google Earth)

99
4.3.2 West Coast of Eyre Peninsula

The Eyre Peninsula, located in APPENDIX I, has a population is around 32,000


(Taylor 2003) and over 40 townships, with the largest being at Port Lincoln at 14,000
(this does not include Whyalla, population over 20,000). The west facing coastline
would receive considerable wave energy, steadily decreasing as you move further up
the coast; whereas the east coast is fetch limited in the Spencer Gulf. Cape Carnot
would receive similar wave energy levels as Cape Du Couedic (slightly less) and is
about 30 km from Port Lincoln, see Figure 4.12. It would be the best location for
supplying wave energy to Port Lincoln, although it appears there is only dirt road
access to the cape and there would need to be grid installed. There would also be
some competition with Hydro Tasmania’s 66 MW Catheral Rock Wind Farm located
north west of Cape Carnot (AusWEA 2005), which is currently under construction.

Figure 4.12 Satellite Image of Cape Carnot (source: Google Earth)

Whidbey Point on the Coffin Bay Peninsula (just north of Cape Carnot) has good
exposure to wave energy and 3 km to the 50 metre contour, but it is ruled out because
Coffin Bay Peninsula is a National Park and it is not located near demand.

100
Further up the west coast there is less electricity demand. As highlighted in the
Bathymetric survey there are many locations where the development small wave
farms could be investigated. Locations with potential are listed as follows:

• The coast adjacent to Lake Hamilton


• Locks Well Beach, between Elliston and Tungketta Reef
• Point Westall and Cape Radstock

A wave farm at Locks Well beach is best located to supply energy to Elliston and
Point Westall is best located to supply energy to Streaky Bay. Again there may be
competition with wind farms at Elliston (Tungketta Hill), were the 120 MW
combined capacity of Stage 1 & 2 has planning approved (AusWEA 2005).

Issues relating to fresh water supplies for the Eyre Peninsula have been a hot topic
over the last few years. Future water demand will soon outstrip supply and limit
economic growth of the region. If the Pelamis could be developed for desalination as
explored in section 5.2 a number of the sites listed above would be attractive locations
for offshore seawater desalination. SA Water commissioned the Eyre Peninsula Water
Supply Master Plan (Taylor 2003) which provides detailed information about the
water resources and demand of the Eyre Peninsula. It proposes a number of scenarios
for desalination including sea water desalination plants, and a 2.3GL/a plant to
desalinate brackish water from the Tod Reservoir at a cost of $32m. The government
also proposes to build a $48.5m pipeline from Iron Knob to Kimba to increase the
Eyre Peninsula’s access to water from the Murray River. Another project known as
“Solar Oasis” based at Whyalla proposes the use of 200 parabolic dish collectors to
generate 24 MW and fresh water at 20 ML per day using multiple flash distillation.
This project has since stalled for a number of reasons, most of them political (despite
strong Whyalla council support) and also due to its “pilot status” and high capital cost
(~$85m). Whatever solution takes place to supply the Eyre Peninsula with fresh
water, there is potential that a Pelamis powered offshore desalination plant could
allow more sustainable use of water and reduce demand on the Murray River. The
supply for the Eyre Peninsula region is limited to about 9.7GL/a (Taylor 2003). Using
the analysis of section 5.2.2, this volume of water could be supplied by about 27
Pelamis located at Cape Carnot desalinating sea water.

101
5 Pelamis Production and Development
Ocean Power Delivery is a manufacturer which has established a network of resources
and contractors in order to produce as many Pelamis units as possible for the
European market. For Australia, Ocean Power Delivery is determined to employ as
many locally established business and resources as possible to manufacture Pelamis.
The majority of a Pelamis can be manufactured in Australia under license with only
some special components needing to be imported from Europe. The benefit of this is
reduced transportation of materials and making use of established local infrastructure
leading to an overall reduction in the cost of projects. Also steel is produced relatively
cheaply in Australia from a readily available supply of high grade ore and abundant
coal resources. The first part of this chapter deals with an overview of manufacturing
Pelamis locally. The second part looks into the potential for using a Pelamis to
desalinate water on an industrial scale. The desalination aspect of the project was not
in the original scope, but has emerged during 2005 as a priority for Wind Prospect.
The proposed work outlined in section 5.2 of this report is at a very early stage and
aims to provide a foundation for further work in desalination by Wind Prospect.

5.1 Local Pelamis Construction

The majority of the cost of a Pelamis can be attributed to the materials and labour
used in producing the structure or hull. Below is a brief description of the process by
which the Pelamis structure is manufactured. Significantly it is very similar to the
process of manufacturing a wind tower, and wind towers are currently being
manufactured in Adelaide.

Steel plate is cut to shape, the edges are bevelled ready for butt welding, the plate is
rolled into a short ring shaped tube and the ends are butt welded together using
multiple pass submerged arc welding. Rings are then welded together end to end to
form long tubular sections. Welding is performed by placing the tube on a platform
with rollers that rotate the tube with a continuous manned welder positioned above.

For the Pelamis, each of the segments would be produced by this process with end
plates and some internal bracing welded in place. Production of the power modules
would be slightly more complex process. As for wind towers, the process would take

102
place on a continuous production line with steel plate building to long tubes
transferred through the plant by roof mounted gantry cranes.

Very little detail has been supplied by OPD regarding the design of the Pelamis,
particularly of the power modules and the dimensions and specifications of internal
components. The present project was to include some effort towards locating local
sources for these components; however this has not been done due to the
commercially sensitive nature of the information required from OPD. The following
list provides an indication of components with potential to be sought or manufactured
locally for an individual Pelamis.

• Generators, 6x 157kVA / 125 kW asynchronous 4 pole ‘squirrel cage’


• Transformer 950kVA step up to 11kV or 33kV
• Accumulators with 350 Bar working pressure and their components
• Rams and Hydraulic Equipment (valves, piping, fittings, turbines/pumps)
• Cable, Chains and tethers used in mooring

Air-Ride Wind Pty. Ltd, based in Kilburn and introduced in section 1.1.2, is well
equipped for the task of manufacturing Pelamis and is enthusiastic about the prospect
of being awarded contracts to build Pelamis wave farms. Segments of the Pelamis
could be transported by rail to locations such as Port Adelaide (or even Portland) for
docking and final assembly. Then individual Pelamis could be towed by a Tug to the
location of the wave farm and then moored to the seabed.

5.2 Developments in Desalination

There is no shortage of water on this planet as the majority of the surface of the planet
is covered by it. However 94% of it contains salt (Buros 2000), greatly limiting its use
to humanity. Moreover, over 90% of all fresh water is locked up in ice at the poles.
From the places where fresh water is available in abundance, currently a large amount
of fossil fuelled energy is used to pump it to locations where it can be stored or
consumed. For example in Australia an enormous amount of power is used to pump
water from the Murray River. In South Australia alone the pumping capacity on the
Murray is 114 MW and the pumping energy used is between 92 and 216GWh per

103
year depending on seasonal rains (Energy SA 2005). Desalination plant capacity
worldwide has been growing steadily since the 60’s (Buros 2000), and has received a
lot of attention in recent years as the demand for fresh water has put great pressure on
supply in many locations of the world. Many cities and consumers are situated near
the ocean, where virtually limitless quantities of water are available. The technologies
to remove salt from seawater are well developed, but they require a lot of energy,
suitable discharge arrangements for the saline concentrate and adequate water storage
reservoirs for the desalinated water.

Renewable energy developers like Wind Prospect are looking beyond electricity
generation to reduce the environmental pressures of a fossil fuelled energy industry.
With the reducing costs of renewable power, the focus is now turning to the
desalination and pumping of water. The mechanical work supplied by renewable
sources such as wind and wave can be applied to desalination using reverse osmosis
and then pumping the water to storage. The following is a list of the main advantages
for using renewable energy to power desalination offshore:

• Given adequate fresh water storage, variability of fresh water production is


not so important. So unlike producing renewable electricity, a lower
capacity factor means fewer limitations to production on a large scale.
• Desalinating water offshore allows salty brine discharge on site rather than
pumping it to shore or elsewhere in the case of a conventional on shore
plant.
• Feed water does not need to be pumped any distance to the inlet of the
desalination plant.
• Renewable powered desalination plants do not need close access to high
power electricity grid, and therefore can be remotely located.

Wind Prospect is currently researching possible performance improvements of reverse


osmosis membranes. Ocean Power Delivery has only just begun to investigate
desalination using the Pelamis but is supporting Wind Prospect with its research. This
section provides a brief overview of desalination and Reverse Osmosis technology
and then investigates how it can be incorporated into the Pelamis design at a
conceptual level.

104
5.2.1 Overview of Reverse Osmosis Technology

There are two main families of desalination technology:


1) The Thermal Processes including Multi-stage Flash, Multiple
Effect and Vapour Compression desalination and
2) The Membrane Processes, including Reverse Osmosis (RO) and
Electrodialysis (Buros 2000).

Other smaller scale technologies include:


- Solar humidification, where direct solar energy is used to
condense water onto transparent surfaces.
- Freezing water to drive out salts.

Of all these technologies the majority of global installed capacity is either Multi Stage
Flash or RO, with about an even share of total capacity. According to the International
Desalination Association (Buros 2000), the growth potential for RO in particular is
tremendous, especially for sea water. Multi stage flash is preferred where waste heat
is available, but RO is more energy efficient in terms of kWh/kL.

The basic concept of RO is to drive salt water through a semi-permeable membrane


under high pressure. The materials of RO membranes are synthetic organic polymers
such as cellulose acetate or polysulfone coated with aromatic polyamides (Sagle &
Freeman 2004). The RO membranes have such tiny pores, diameters of 1 to 10
angstroms (10-10 metres, or 10-4 microns), that salt ions will not pass through them
(Sagle & Freeman 2004). The basic components of a RO plant are shown in Figure
5.1. The pretreatment is important to keep the RO membranes clean and to prevent
microbial growth and is usually achieved by fine filtration to remove suspended
particles and the addition of chemicals to kill microorganisms. High pressure pumps
are needed to drive water into the membrane assembly, a vessel that allows water to
be pressurised against the membrane beyond its osmotic pressure. Sagle & Freeman
(2004) indicate that the osmotic pressure of a conventional membrane for sea water of
salt concentration 32g/L, is 23 bar at 25°C (2004), whereas Buros (2000) suggest that
in practice the pump needs to apply more like 54 to 80 bar to produce a reasonable
flow rate.

105
Figure 5.1 Basic components of a RO plant (Buros 2000)

Exiting the membrane assembly is the fresh water or permeate and the salty
concentrate discharge or brine. No membrane is perfect and small concentrations of
salt at typically about 1% of the feed water concentration remain in the product water.
Post treatment involves stabilizing the water, including removing gases and adjusting
the pH levels.

There are four main types of RO modules:


i. Plate and Frame
ii. Tubular
iii. Spiral Wound
iv. Hollow Fibre.

The most popular configurations for industrial sea water desalination are Spiral
Wound and Hollow Fibre. They have the highest surface area per unit volume
allowing for the most compact design, as would be needed for use in the Pelamis. The
design of each configuration can vary widely between manufacturers, but an example
of a Hollow Fibre membrane assembly is illustrated in Figure 5.2.

106
Figure 5.2 Hollow Fibre RO membrane configuration (Buros 2000)

Importantly, the reject brine leaves the RO assembly still under high pressure,
typically only 1 to 4 bar below the feed pressure (Buros 2000). It is becoming more
common in RO plants to design efficient recovery systems for this energy by the
means of a turbine, pressure exchanger or “hydraulic turbocharger”. How effectively
this energy can be captured and delivered back to the feed pump can have a major
influence on the whole energy efficiency of a RO plant. For applications in renewable
energy, such as the Pelamis, energy efficiency is critical to the overall performance of
the design.

The main advantage of RO compared to Multi-Stage Flash desalination or other


thermal processes is that no heating is required, leading to the highest energy
efficiencies for desalination processes. Oceanic brine discharge from RO desalination
can have a detrimental effect on water ecosystems by reducing water oxygen content.
In the case of the Pelamis, discharge would occur at depths greater than 50m,
relatively displaced from shoreline marine life. Environmental impacts can also result

107
from RO desalination by the discharging of chemicals with the brine that were added
in the pretreatment stage. In the case of the Pelamis, which is promoting
environmentally friendly design, effort may need to go towards eliminating harmful
chemical additives from the brine discharge. This has already been done for the
Penneshaw plant at Kangaroo Island (Kirke pers. comm). This may only come about
from greater investment in filtration equipment at the pretreatment stage, which would
bring about improvements in the membrane performance overall.

RO has been the most rapidly advancing form of desalination technology in recent
times. Improvements in the efficiency and durability of membranes, energy recovery
and efficiency, membrane life and operational experience have led to an overall
reduction in the cost of fresh water produced from saline water. Unit costs around US
50c/kL are now being quoted (Kirke pers. comm). This would make RO competitive
with traditional water sources in Australian cities. However the chemical stability and
fouling of membranes are areas where significant improvement could be made. With
its current research program, Wind Prospect aims to significantly improve the
efficiency of RO membranes.

108
5.2.2 Pelamis Design Integration

Conventional RO plants are located in buildings and have all pumps driven by
electrical power. The membrane units are subjected to a constant pressure with
minimal fluctuation at or near capacity. Osmoflo is an Australian company that
manufactures RO membranes for desalination and was supportive in providing some
information about their desalination equipment. Based on an average generation of
300 kW (7.2 MWh per day) from a single Pelamis (Capacity Factor = 40%),
according to Osmoflo each Pelamis would be theoretically capable of producing 1
megalitre of fresh water per day (MLD) on average from sea water and discharge
approximately 32,000 kg of salt per day. This gives 7.2 kWh/kL, whereas the best
large plants need about 4 kWh/kL (Kirke pers. comm). Osmoflow indicate that the
capital cost of desalination equipment to produce 1MLD is in the order of $1.5m.

In order to provide a stable input of pressure to the membranes, enough to exceed the
osmotic pressure, there is a necessity to retain the accumulators in the power modules.
Energy from the waves must be stored between waves (period typically 7-8 sec) to
even out the fluctuations. The working pressure for the Pelamis accumulators is 150
to 300 bar, much higher than the sea water osmotic pressures of up to 80 bar
mentioned earlier. This leaves three possible configurations for driving RO pumps
using Pelamis power as follows:

1) Electricity is generated using the existing design and then fed to electric
pumps for RO.
2) Hydraulic fluid (oil) exiting the accumulators drive turbines to power
generators in the existing design, but the turbines could be used instead to
power feed pumps for RO.
3) Sea water could be pumped by the rams directly into accumulators
which could directly supply pressure to the RO assembly.

Ideally the cost of desalination could be reduced by eliminating all electrical


components and having a purely mechanical system provide the work for RO. High
power and voltage are usually accompanied by expensive equipment, such as
generators, capacitors and transformers and there are also the losses in generators
powering motors. The main advantage of using electric power as in case 1) is that

109
energy can be easily controlled and diverted. This may prove beneficial in the case of
a number of Pelamis in a wave farm powering desalination explored later.

Case 2) explores a purely mechanical system which makes use of most of the existing
hydraulic equipment. The merit in this concept is avoiding sea water coming into
contact with all but the rods driven by the Pelamis segments and the feed pump, to
avoid corrosion issues and possible scaling, although this may be an unnecessary
design. Case 3) is simpler except all of the surfaces that come into contact with sea
water will need to be corrosion resistant. There would already be a range of saline
resistant hydraulics employed in RO plants. Some surfaces could be made of
polyurethane and steel cylinders could have non-corroding and abrasion resistant
lining. Accumulators generally have a rubber bladder which separates the pumped
water from the compressed gas (nitrogen) and there seems no reason why the sea
water could not be contained within the rubber bladder. An advantage of Case 3) is
that a turbine or pump is not needed to pressurise the RO assembly, only a valve or set
of valves to regulate the pressure leaving the accumulator. In either case 2) or 3),
some means is needed to recover the pressure from the reject brine as it leaves the RO
assembly. In case 1) and 2) the pressurised brine could be used to boost the feed
turbine. In any of the cases the pressurized permeate could be used to assist the
hydraulic rams by some arrangement while also providing energy to pressurise pre
treatment filtration.

There are a few options regarding the location of the RO assemblies inside a Pelamis.
Osmoflo indicates that a 500KLD (kilolitre day) RO unit can be containerised into a
40ft shipping container which has a volume of about 31 cubic metres. The dimensions
of a the structure of a Pelamis power module are about 3.5m diameter by 3.5m long,
giving a capacity of about 33.6 cubic metres. This means that a 500KLD unit may just
fit inside a power module if all of its contents were removed. Given there are 3 power
modules it may be possible to fit a ~333KLD unit in each module. Either that or the
main segments of the Pelamis could be used where there is much more space. Perhaps
the two inner segments could each contain a 500KLD unit or larger. Power from the
two outer modules could be fed inwards and the inner module could be split power
each or either way. The nose segment of the Pelamis could be used for intake and
contain all of the pretreatment equipment and feed all of the desalination throughout

110
the Pelamis with discharge at the tail. It is uncertain how the mass requirements of
1MLD RO desalination would affect the buoyancy and weight distribution of the
Pelamis. Ideally the tonnage and distribution of sandbags used as ballast could be
matched. In practice an overall RO capacity of 1 MLD would be used because this is
only the average output and a capacity of up to 2 MLD may be required to make best
use of high energy periods. The research by Wind Prospect aims to increase the
efficiency of the membranes and so may reduce the overall volume requirement.

An important consideration with using the Pelamis to operate RO desalination is the


variability of energy supplied by waves. It may be just a case of production starting
once osmotic pressure has been obtained and then pressures growing well beyond
osmotic pressure under normal operation at part capacity. There may be a limit to the
rate of membrane permeation. Power available at the hydraulic rams is below mean
output for the majority of the time. So there may need to be a balancing of the
distribution of power to ensure all operating RO units are supplied with pressures
above the osmotic pressure at times of low available energy, to eliminate wasting
power on idle units. With case 1) where electricity is generated, the power could be
supplied to individual units on a cycle at part load and to all units at full load. Using
the purely mechanical processes it may be feasible to divert high pressure water from
section to section within the Pelamis. It may be that each RO unit within a Pelamis, be
it 3 or less, could be divided into portions or a number of separate sub assemblies
running in parallel. The sub assemblies could be cycled using valves at times of low
available energy.

It appears that an entirely mechanically driven Pelamis desalination device could be


feasibly engineered by a number of different combinations. There are cost benefits in
having fewer units and components per Pelamis if possible. A desalination capacity
would need to be found which balances the use of high energy periods and operation
during the more frequent periods of low energy. Only a detailed analysis could result
in identifying the best solution taking into account factors such as equipment cost,
durability, maintenance and overall performance and efficiency. The following
section will explore the operation of an offshore Pelamis desalination plant and some
other emerging concepts.

111
5.2.3 Offshore Desalination Plant Concepts

A Pelamis wave farm with integrated RO units is likely to produce around 1MLD or
11.6 litres per second on average per Pelamis at a Capacity Factor of 40%. As soon as
the fresh water is produced it will need to be pumped to storage, but in any case it will
need to be pumped to land so it can be used. The energy required for this pumping
depends on the size of the plant, the pipe diameter and the distance to shore but is
likely to be a small proportion of the energy used in desalination. Flexible hoses from
each Pelamis could meet at a moored floating junction nearby where a dedicated
electric motor pumps water down a single pipeline that runs to the shore. Only a small
amount of electrical energy would be needed from the Pelamis to power this motor if
the desalination units are designed to add some pressure to the product water after
energy recovery. Post treatment (see Figure 5.1) of the water could take place on land
where the plant and the water quality can be better monitored.

One idea is to use an electricity generating Pelamis wave farm to power an offshore
desalination platform or vessel that pumps the water to shore. An old ship or
decommissioned tanker could be used not only for a desalination plant but as an
intermediate storage vessel for fresh water. This vessel might be used to tow the
Pelamis to shore for maintenance or equipped as a platform for offshore maintenance.

An innovative yet radical concept is being explored by an Australian “innovation


architect” John Dobozy known as the Aqudam (Pyper 2004). The concept involves a
giant 3 km long and 1 km wide flexible membrane acting as a floating water storage
reservoir for rain water, storm water and desalinated sea water. It is supported by a
number of floats and anchored to the seabed by a number of pylons, expanding as it
fills. The dam would be divided into a number of compartments containing different
water sources and insuring against leakages. The dam is targeting locations of cities
by the sea where currently large amounts of fresh water are collected and fed into the
ocean during storms. This project is still at a conceptual stage with many engineering
challenges to be explored but central to the desalination aspect is the use of renewable
energy. A structure such as the Aquadam could complement a Pelamis desalination
plant and using Pelamis power to clean storm water.

112
Figure 5.3 The Aquadam concept (Pyper 2004)

Offshore wind power is a direct competitor of the Pelamis and there is research
underway that is directed at desalination by wind energy. Wind DeSalter Technology
(WDE 2005) proposes to use offshore wind turbines to pump high pressure water
down RO units located within the tower by mechanical process. Again, this concept is
only at a conceptual level but given the growth of wind farms and falling capital costs,
Wind DeSalter could soon become economically competitive for large scale
desalination. A key advantage of offshore Pelamis desalination could be that
individual Pelamis could be towed to a port for maintenance and servicing whereas
offshore wind must be operated on out at sea. Another advantage is that the Pelamis
would have significantly lower visual impact lending to a smaller chance of public
objections to developments. Also the Pelamis is not likely to kill a single sea bird,
even if wind turbines very rarely do so.

113
6 Life Cycle Analysis
A life cycle analysis (LCA) is a method of assessing the environmental aspects of a
product and its potential impacts and consequences. This means looking at the entire
life cycle from producing raw materials right through to disassembly for recycling.
For a renewable energy project which aims to be ecologically sustainable, it is
desirable to minimise adverse environmental impacts. A LCA is a very useful tool for
monitoring the environmental consequences of a renewable energy project and indeed
comparing it to conventional fossil fuelled energy. Financial and social factors are not
included in an LCA but it is a powerful tool for measuring the “true” cost of a product
or project to the environment and ultimately society. OPD declined to provide a LCA
for the Pelamis so it was decided to undertake a very basic analysis to get an idea of
how it performs and compares with other technologies. A LCA is not designed to
glorify the destructive aspect of a renewable energy project but should highlight the
attractiveness of renewable energy technologies both ecologically and economically
next to conventional energy production.

A highly detailed LCA is a complex project requiring a range of information


resources and has some inherent limitations. These include making many assumptions
and selections, and generating models which may not hold for all circumstances.
Often data are very hard to access and authenticate, or does not even exist.

6.1 Embodied Energy for a Locally Manufactured Pelamis

An important component of a LCA for renewable energy is to determine the


embodied energy of a project. This is the total amount of energy required to make the
product from raw materials. In most instances this energy is currently derived from
non-renewable sources. Embodied energy is a potentially huge area of research that
has been greatly unexplored until recent times. This is because the research is time
consuming and expensive, and is often perceived as having little economic benefit to
business. Some may even find it disturbing to explore the LCA and discover the
embodied energy requirement of a new car, house or consumer product.

114
The following sections will attempt a very basic exploration of the LCA for a single
Pelamis within a wave farm, using the limited information supplied by OPD. Focus is
put on the embodied energy requirement, stating all assumptions. The embodied
energy of a Pelamis unit would only be slightly reduced by mass production. In many
cases data was not readily available and so each of these aspects will only be treated
as points of discussion for their contribution to the LCA.

It is assumed that the great majority of the total mass of a Pelamis, without sand
ballast, would be comprised of steel, perhaps over 95%. The rest of the mass would
include the hydraulic fluid, copper (in the electrical components: generators, cables),
rubber or plastic (seals, insulation) and even red paint. Using the geometries of the
Pelamis calculations can be made to estimate the total mass of steel. The prototype is
120m long and made of 25mm plate and the production model was indicated by OPD
to be 150m long and 3.5m diameter, but made of 20mm plate. For this analysis only
the production model will be considered. Assuming that the density of the steel is
7850 kg/m3 the mass is calculated by determining the volume of steel as follows:

Cylinder _ volume = 150 × 3.5π × 0.02 = 33m 3


Disc _ volume = 13 × π (3.5 / 2) 2 × 0.02 = 2.5m 3 (6.1)
Structural _ mass = (33 + 2.5) × 7850 = 278,675kg

The disc volume represents the 13 circular steel plates used on the ends of each of the
segments and is assumed to be of the same thickness as the cylinder plate. The mass
of the yoke and the internal steel structuring is more difficult to calculate but an
estimate is made to assume they would bring the unit steel mass up to 300 tons.

There is also a significant mass of steel used in mooring the Pelamis including a tether
weight of large gauge chain (about 40 tons was indicated), a clump weight each side
weighing 35 tons each (mostly steel according to its dimensions) and a set of
embedment anchors (4 for a single Pelamis). A multi Pelamis wave farm has each unit
sharing clump weights and embedment anchors with surrounding units. Photos of
these components and diagrams of mooring arrangements could not be shown for
reasons of commercial confidentiality. The mooring cable is another consideration; it
is 83mm in diameter and made of multiple steel wire strands similar to suspension

115
cable. There is at least 50m of it hanging beneath the Pelamis to the tether mass on the
sea floor and four lengths, which appear to be over 100m in length running to each
anchor for a single Pelamis unit. This total length is assumed to be about 500m giving
a total steel cable volume of about 2.7 m3 and a mass of about 20 tons. An actual
value for the total steel mass of these mooring components was not provided. So a
generous estimation can be made to assume for this analysis that the additional steel
mass for mooring a Pelamis unit within a wave farm is (40 + 35 +2*20 +20) 100 tons.

Determining the embodied energy of steel is a complex process and requires an


understanding of the steelmaking process. The main raw resource inputs to a modern
integrated steelworks are Iron Ore (usually hematite Iron (III) Oxide - Fe2O3), Coal
and Limestone. Most of the coal is made into Coke, which is simply coal that has
been cooked in an oven so that it is ready to be fed into the Blast Furnace. Air Ride
uses steel manufactured by Bluescope, Australia’s largest steel company, to
manufacture its wind towers. For the sake of this analysis it is assumed the locally
manufactured Pelamis would source its steel from Bluescope’s integrated steelworks.

Figure 6.1 Energy used by Bluescope to make steel (Bluscope 2005)

As shown in Figure 6.1, over 90% of the energy used to produce steel from raw
materials comes from coal combustion. There would be also considerable amount of
energy (mostly oil derived) needed to mine the iron ore, coal and limestone and then
transport it to the steelworks. Resultant figures depend on where the boundaries are
drawn in the analysis, because a number of factors could be included such as energy
needed to operate and maintain the factories and offices, and transport for workers.
Embodied energy is usually a measure of the Process Energy Requirement (PER), that

116
is the energy directly related to the manufacture of the material (Milne 2005). The
Gross Energy Requirement (GER) is a true indication of the embodied energy and
encompasses very wide boundaries and is usually impractical to measure (Milne
2005). In most cases the PER is 50 to 80% of the GER. It is worth mentioning that
producing steel requires a large amount of fresh water, but in most modern steelworks
this water is recycled.

The resources quoted in APPENDIX J provide an indication of the range of


estimations made for the PER of steel. In general, steel that has been minimally
processed has an embodied energy of 35 MJ/kg. The majority of this value is made up
of the energy content of the coal used. More processing results in a higher embodied
energy. For example the higher figure estimated by Buchanan of 59 MJ/kg for
structural sections is due to the reheating of steel blooms before they are rolled.
Bluescope claims that “The resource energy required to extract and refine iron ore in
the ground to produce steel is approximately 26 gigajoules/tonne (Bluescope 2005)”.
Importantly, the life cycle of steel is continuous because it is 100% recyclable,
meaning almost all of the Pelamis is recyclable. Due to the high value of steel, it is the
most recycled material in the world and scrap steel is actually needed to make new
steel. Bluescope claim that 15 to 20% of their steel is sourced from scrap (Bluescope
2005). A figure for recycled steel sections is estimated by Alcorn of 8.9 MJ/kg in
APPENDIX J. This is significantly lower than 35 MJ/kg and the 26 MJ/kg quoted by
Bluescope may incorporate the recycled component.

This analysis gives an indication of how difficult the estimation process for steel
embodied energy is. It gets even more complicated when materials are subjected to
additional manufacturing process to produce machinery. For the Pelamis, the most
obvious additions to the embodied energy of the structure from manufacturing is the
welding, shaping, cutting, scrap and operating the factory cranes & lighting etc.
Pistons, accumulators, generators, and the mooring cable and equipment would also
have considerable additional embodied energy from manufacturing. Only a thorough
investigation could result in an accurate estimate, but for the sake of this analysis a
guess of 10% (could be more) is added to the 35 MJ/kg to give 38.5 MJ/kg for the
structure of the Pelamis. Using these assumptions producing a single Pelamis
structure, assumed at 300tons, would have an embodied energy calculated as follows:

117
300,000 × 38.5 = 11,550,000MJ = 3,208MWh (6.2)

This figure can be considered as the embodied energy of a Pelamis unit. The true
embodied energy would be higher this when the GER is considered and that
manufacture of all the components are included. Including the assumed 100 ton steel
mass of the mooring equipment, which would be minimally processed (except the
multi-strand mooring cable), the embodied energy is recalculated for a single installed
Pelamis as:

100,000 × 35 + 11,550,000 = 15,050,000MJ = 4,181MWh (6.3)

These figures do not include transportation and installation, so a few more


calculations are made to get an idea of how this would contribute to the total
embodied energy. Firstly some more assumptions are made for the sake of
calculation. The wave farm under analysis is located at Cape Du Couedic, Kangaroo
Island, with Pelamis manufactured in Adelaide and steel sourced from Port Kembla,
NSW. Using the figures provided by Lawson (1996) in APPENDIX J the figures for
transportation are assumed for rail 0.5 MJ/ton.km and ship 0.12 MJ/ton.km. The
distances by rail from Port Kembla (Wollongong) to Adelaide is assumed to be
1500km and the shipping distance from Port Adelaide to Cape Du Couedic as 250
km. The energy required to transport the full 400 tons of steel is calculated as follows:

Rail = 0.5 × 400 × 1500 = 300,000 MJ = 83.3MWh


Ship = 0.12 × 400 × 250 = 12,000 MJ = 3.3MWh (6.4)
Total = 83.3 + 3.3 = 86.6MWh

These calculations demonstrate how small the transportation energy requirements are
relative to the steel manufacturing energy, at about only 2% in this case. The energy
needed to transport an assembled Pelamis from the UK would be considerably more.
Note that these calculations ignore the mass of the sand ballast of about 400 tons (a
ballasted Pelamis weighs 700 tons), which would raise the transportation costs
depending on its source location. Sand is assumed to have negligible embodied
energy.

118
The submarine cable would have considerable embodied energy in its manufacture,
transportation and installation. An assumption is made that the cable is 5km long and
is rated at 33kV and can carry a capacity of 30MVA (909 Amperes) for a 40 Pelamis
wave farm. Some specifications were found for underground 33kV cables (Eland
2005) to get an idea of the PER of the submarine cable. A cable with a suitable
continuous current carrying capacity of 940 Amperes would have an overall diameter
of 71.6 mm and a nominal weight of 9150 kg/km, giving a total cable mass of about
46 tons. This cable has a copper conductor (although aluminum is also common) and
is sheathed in PVC and insulated in XPLE (cross linked polyethylene) plastic.
According to Lawson (1996) the embodied energy of unprocessed copper is 100
MJ/kg, PVC is 80 MJ/kg and plastic in general is 90 MJ/kg. So an assumption is
made that the embodied energy needed to make the cable would be about 100 MJ/kg
when manufacturing process energy is included, giving a total figure of:

46,000 × 100 = 4,600,000 MJ = 1,278MWh (6.5)

Transportation and installation energy would add to this figure depending on


circumstances, but from the calculations made earlier (equation 6.4) it is unlikely be
to a large component of the manufacturing energy. Now dividing this figure up among
40 Pelamis reduces the figure to only 32 MWh per unit, which is less than 1% of the
overall embodied energy.

A rounded figure for the overall embodied energy of an installed Pelamis unit is
presented in equation (6.6) using equations (6.3), (6.4) and the 32 MWh for the cable.
Using all the assumptions stated in this analysis, this value represents the PER, or the
energy required in directly manufacturing and installing a Pelamis. The bulk of it is
from the mass of steel which needs coal for the blast furnace and basic oxygen
steelmaking. This figure is only an estimate and there is the possibility that it contains
significant errors and omissions that could only be eliminated by a more thorough
analysis.

4181 + 86.6 + 32 ≈ 4,300MWh ( ≈ 15,500,000MJ ) (6.6)

119
6.2 Energy Balance

The energy payback period for the Pelamis is the amount of time taken for the energy
produced by the Pelamis to meet the amount of energy invested or embodied energy
of the project. It is assumed that a single Pelamis unit (as calculated in the previous
(section) would have a payback period only slightly larger than that of a large wave
farm. As with the previous section the Pelamis is assumed to be located at Cape Du
Couedic (for the sake of this analysis) with an annual capacity factor of 40%. The first
step is to calculate the annual electrical energy produced by the Pelamis.

0.75MW × 0.4 × 24 × 365 = 2,628MWh per year (6.7)

Equating the produced electrical energy directly to the embodied energy gives the
following payback period:

4300 / 2628 = 1.636 years (6.8)

Given that external engineering firm WS Atkins has verified the design for a life of 15
years; the Pelamis only spends about 11% of its life cycle in energy debt. Using the
capacity factor of 50% for Cape Sorell the payback period is only 1.31 years, or 8.7%
of the verified life cycle. After this period of debt the Pelamis is generating energy for
free (disregarding maintenance energy) and may indeed last longer than 15 years.

However electrical energy is a highly valuable form of energy that always involves
thermal losses when sourced from fossil fuels. Another way of looking at the payback
period is to look at how long it would take the Pelamis to generate the electrical
energy otherwise generated by the amount of coal used to make the Pelamis. The
following assumptions are made for this calculation:

• Coal represents the entire source of the embodied energy of the Pelamis,
even though it would only make up the majority.
• 100% thermal efficiency of coal energy content used in making the
Pelamis, even though the thermal efficiency of steelmaking would
certainly be less than 100% (data could not be sourced).

120
• The efficiency of electricity production from coal using a steam cycle
turbine is 36%; the value achieved by the coal fired Bayswater Power
Station, NSW (MacGen 2004).

The payback period can then be calculated as follows using the annual energy
production from equation (6.7) with a capacity factor of 40%:

1.636 × 0.36 = 0.59 years (or about 7 months) (6.9)

This is an even more favorable payback period comparing the use of a given amount
of coal used for generating electricity directly or for making a Pelamis to generate
electricity. Yet another payback period could be calculated in the same way
accounting for how long it would take the Pelamis to pay back the amount of carbon
dioxide generated. This was not calculated due to insufficient data for emissions in
steelmaking and for coal combustion (there are many different grades of coal).
However it can be imagined that the payback period would be similar to that
calculated above.

Reducing the embodied energy of the Pelamis would mean reducing the weight,
possibly to the detriment of strength, or selecting different materials. Using 20 mm
plate instead of 25 mm plate on the prototype does make quite a difference. Making a
lighter Pelamis out of fiberglass for example may reduce its embodied energy, but
may not be a major improvement when recycling is considered.

A figure is claimed in the OPD brochure in APPENDIX A for the Pelamis saving
2000 tonnes CO2 emissions per year for electricity otherwise produced from a gas
turbine power plant. Although it is not specified what capacity factor the Pelamis
operates at, this is a very significant saving given the short energy payback periods
found in this analysis. Even more so when you consider that the Pelamis (once
manufactured) will not release other harmful emissions such as carbon monoxide,
nitrogen oxides and particulate matter produced by fossil fuelled power plants.

121
6.3 Conclusion

Regarding the environmental impact assessment of a site, OPD has provided a generic
scoping study located in APPENDIX K. This deals mostly with the onsite
construction and operation of the wave farm and lists a range of potential
environmental impacts. The direct impact would be the submarine cable and mooring
equipment on the sea bed and the onshore cable arrangement. Depending on location,
a wave farm could disrupt shipping routes, fisheries and recreation. The Pelamis
generates some undersea noise which could disturb cetaceans. If a very large wave
farm was installed that extracted a lot of wave energy, the size of the waves arriving
at the shore would be reduced to some degree, but probably very minor. This is
unlikely to have major impacts on a coastal marine life, but this could only be verified
by a specialized study. The visual impact of a Pelamis is minimal from the shoreline
and much less than an offshore wind farm. In order to obtain the relevant licensing
and consents, an environmental impact assessment report must be produced for a
particular site.

While it is true that a wave farm requires a great amount of steel and embodied
energy, the payback periods are very short with respect to the whole life cycle. It
would be shorter than that of Energetech’s WEB (see section 2.3.2) which requires
485 tons of steel with a more sophisticated structure and has a smaller claimed output
capacity of only 500kW. The life cycle analysis of Vestas offshore wind turbines were
analysed for comparison to the Pelamis. Again a great number of assumptions were
made for the analysis but the payback period for a 2 MW V80 turbine is ~9 months
and a 3 MW V90 turbine is ~6.8 months (VWS 2005), compared to 1.6 years for a
Pelamis. In general the period shortens for larger turbines and is slightly shorter
onshore than offshore. The capacity factors used were generally 30% and the life span
of a turbine is considered to be 20 years. Ideally a 750 kW turbine should be used for
comparison, but no data was available.

Overall it could be said that if a Pelamis lasts 15 years or more then environmental
benefits of using the Pelamis to generate electricity, compared with conventional
sources, greatly outweigh its impacts.

122
7 Conclusions and Recommended Further Works
A conclusion reflecting the title of this project would be that the wave energy
potential for Kangaroo Island is not particularly good due to a combination of the
island’s geography, directional wave climate and grid infrastructure near demand. It
would not be economically feasible to develop a Pelamis wave farm to substantially
meet the energy demands of the island. Some alternative renewable energy solutions
for the island are discussed in section 4.1.4. This led to setting a much broader aim
and scope for the project of which almost every aspect was successfully addressed
apart from those which had restricted access to information. These include sourcing
local components and a financial analysis. The project was able to identify the best
sites for wave energy in southern Australia and asses their feasibility, so achieving the
general aim of the project.

Portland in Victoria was found to be the most promising location for a Pelamis wave
farm in southern Australia. Cape Duquesne and Cape Nelson receive the most
abundant wave energy resource on mainland Australia with ideal coastal bathymetry
for the Pelamis. It is estimated that a single Pelamis at this location could generate a
mean annual electricity output of 2.9 GWh at a capacity factor of 44% (330 kW).
Unlike a wind turbine it was found to be very rare that the Pelamis would be idle or
generate the full 750 kW. At both sites the submarine cable would only need a length
of 1.5 km to meet the 50 metre depth contour, a condition only found at a few sites
around Australia. Powercor, Victoria’s largest electricity distributor, are interested in
helping Wind Prospect determine the feasibility of wave farm connection to the
Portland substation. Australia’s largest wind farm is proposed for Portland and might
be a source of competition. This could be met by the observed advantage that wave
energy is usually still strong when there are sudden drops in wind energy, which can
be frequent at times (see Figure 3.14).

It is recommended that modelled NOAA WW3 data be purchased by Wind Prospect


for further and more accurate investigation into the wave resource at Portland.
Currently Wind Prospect is not prepared to do this, partly due to the decision that it is
not the correct political environment to develop wave farms in Australia. At the
moment Wind Prospect are more committed to researching renewable desalination. It

123
is also recommended that further research goes into determining the costs of
submarine cables and offshore water pipelines for local projects, and in finding ways
to reduce their capital and installation expenses.

Other locations which were found to have potential for wave energy projects include
the west coasts of Tasmania and the Eyre Peninsula, South Australia. In each case the
most promising sites are located some distance from loads and there would need to be
some development of grid infrastructure. Many of these sites were also chosen to
develop wind farms, which may result in some competition, but may also be an
advantage in sharing newly developed infrastructure. It is recommended that further
investigation goes into the feasibility of using wave energy at some of these sites.
Fresh water supply issues were explored for the Eyre Peninsula and there is potential
that offshore Pelamis desalination of sea water could provide an attractive solution.

A conceptual investigation was made into using the Pelamis for desalination. It
proposes a number of configurations for incorporating reverse osmosis units and
consideration for plant design. There appears to be no reason why a purely
mechanical means can’t be used to convert the pumping power of the Pelamis to
reverse osmosis desalination within its structure. In this way expensive electrical
equipment can be avoided. It was estimated that a single Pelamis located at the
southern tip of the Eyre Peninsula could produce about 1 megalitre of fresh water per
day from sea water. In this case about 27 Pelamis could meet the supply capacity of
the whole region at 9.7 GL/a.

A brief life cycle analysis (LCA) was undertaken for the Pelamis with a focus on the
embodied energy of the design. Using a lot of assumptions it was calculated that a
Pelamis would need about 4,300MWh of energy to produce, mostly for the 400 tons
of steel required. It has a direct energy balance or payback period of about 1.6 years,
which is small considering its 15 year design life. This figure is reduced by up to 20%
if concrete were used in the mooring weights. While shorter than other wave energy
converters, such as Energetech, this payback period is significantly longer than for
wind turbines. It is recommended that OPD undertake a detailed LCA and make it
freely available, as done by Vestas Wind Systems.

124
According to Wind Prospect, the main value of this project is the resource assessment
work specific to southern Australia. This includes a detailed study of wave energy
resources and a bathymetric survey of the entire south Australian coastline. Statistical
analysis was undertaken for 10 years worth of combined wave rider buoy data from
Cape Du Couedic and Cape Sorell, supplied by the Bureau of Meteorology. A
summary of some major findings for the wave energy resource study are listed as
follows:

• Long term mean wave energy is 33 kW/m at Cape Du Couedic, Kangaroo


Island and 42 kW/m (about the highest in Australia) at Cape Sorell,
Tasmania.
• Long fetch swell directional climate to southern Australia is westerly to
south westerly. Wave energy is high when there are strong winds from
these directions.
• A shift in wind direction from northerly to westerly results in an abrupt
increase in wave energy along with a cool change.
• There are distinct seasonal trends with low energy in summer months, high
energy in winter months but usually peaking in September. There are
considerable year to year variation in monthly means.
• Wave heights and available power show skew to the right distributions,
whereas wave periods are close to a normal distribution.
• Mean day to day variation in daily mean wave energy is about 50% of the
annual mean value.
• Southern Australia receives some of the highest wave energy levels in the
world, rivalled or exceeded only by NZ, southern Chile and the UK
(according to the ERA-40 model)

125
References

Attenborough, D 1979, Life on Earth, Fontana/Collins, Glasgow.

AGO 2003, Renewable Energy Commercialization in Australia, Australian


Greenhouse Office, Canberra.

AusWEA 2005, Wind Farms in South Australia, Australian Wind Energy


Association, viewed 2 November 2005.
<http://www.auswea.com.au/projects/sa.asp>

Beyen A & Wilson JH 2005, Wave Energy Capacity and Part-Load Operation, Sea
Technology magazine, July 2005 edition.

BlueScope 2005, Energy Use in BlueScope Steel’s Steelmaking, BlueScope Steel,


viewed 24 October,
< http://www.bluescopesteel.com/corp/navajo/display.cfm/objectID.9DC1B6BE-
94A7-40CD-9C7253831292C019>

Brooke J 2003, Wave Energy Conversion, Elsevier Ocean Engineering Book Series,
Vol 6, Elsevier.

Boyle, G (ed.) 2004, Renewable Energy: Power for a Sustainable Future, Oxford
University Press, Oxford.

Buros, OK 2000, The ABCs of Desalting – Second Edition, International Desalination


Association.

Chahine R & Bernard P 2001, Assesment of Hydrogen Storage on Different Carbons,


IEAHP (International Energy Agency Hydrogen Program), viewed 22 May 2005,
<www.ieahia.org/pdfs/NN0094v.pdf>

126
DEH (Department of Environment and Heritage) 2005, Human Settlements: Energy
Trends, viewed 22 May 2005.
<http://www.environment.sa.gov.au/reporting/human/energy.html>

Duckers, L 2004, ‘Wave Energy’ chapter 8 in – Boyle G (ed.) 2004.


Eland 2005, 33kv Cable: Energy Characteristics, Eland Electrical Cables, UK,
viewed 2 November 2005.
<http://elandcables-specs.co.uk:8080/eland/specification?CableTypeID=139>

Energetech 2005, – various titles, Energetech Australia Ltd.


<http://www.energetech.com.au>

Energy SA 2005, Water Energy, Energy SA Advisory, viewed 20 October 2005,


<http://www.sustainable.energy.sa.gov.au/pages/advisory/renewables/types/water/wat
er.htm:sectID=21&tempID=52>

ESIPC 2004, Annual Planning Report – June 2004, Electricity Supply Industry
Planning Council, Adelaide.

ESIPC 2005, Annual Planning Report – June 2005, Electricity Supply Industry
Planning Council, Adelaide.

Fay JA & Golomb DS 2002, Energy and the Environment, Oxford University Press,
Oxford.

Hayden, HC 2001, The Solar Fraud, Vales Lake, Pueblo West.

IEA (International Energy Agency) 2000, International Energy Outlook 2000,


OECD/IEA, Paris.

IEA (International Energy Agency) 2002, Beyond Kyoto, Energy Dynamics and
Climate Stabilisation, OECD/IEA, Paris.

127
IEA (International Energy Agency) 2004, Key World Energy Statistics – 2004
Edition, OECD/IEA, Paris.

IPC (International Programs Centre) 2005, World POPClock, U.S. Census Bureau,
viewed 4 November 2005.
<http://www.census.gov/ipc/www/popclockworld.html>

IRN (International Rivers Network) 2005, Three Gorges Campaign, viewed 1 April,
2005. < http://www.irn.org/programs/threeg/>

Irvin R, Thiagarajan KP & Thomas M 2005, The Hydrodynamic Efficiency of an


Oscillating Water Column, 24th International Conference on Offshore Mechanics and
Artic Engineering, Halkidiki.

ISA (International Society of Aboriculture) 2005, Fun Facts about Trees, viewed 27
March, 2005.
<http://www.treesaregood.com/funfacts/funfacts.asp>

Johansson, T. et al. (eds) 1993, Renewable Energy: Sources for Fuels and Electricity,
Earthscan, London.

Kaiser KLE 1999, Hydrogen, Fuel of the Future?, Canadian Chemical News, viewed
16 August, 2005
<http://www.terrabase-inc.com/Hydrogen.pdf>

Kirke B 2004, KI Energy Supply Options, KIDB/UniSA SEC, Adelaide.


<http://www.kangarooisland.org/reports/KItripreportsept04.pdf>

Kordesch et al. 2003, Ammonia as Hydrogen Source for Alkaline Fuel Cell-Battery
Hybrid System, Graz Technical University, Austria, viewed 21 October 2005.
<http://www.zapworld.com/about/news/Ammonia%20as%20Hydrogen%20Source%
20Paper%20Nov%2011%2003.pdf>

128
Kruse B, Grinna S & Buch C. 2002, Hydrogen - Status and Possibilities, Bellona
Foundation, viewed 1 November 2005.
<http://www.bellona.no/data/f/0/26/97/0_9811_1/Hydrogen_6-2002.pdf>

Lawson, B 1996, Building materials, energy and the environment : towards


ecologically sustainable development, Royal Australian Institute of Architects, Red
Hill, A.C.T.

MacGen 2004, Bayswater Power Station, Macquarie Generation, NSW, viewed 2


November 2005.
<http://www.macgen.com.au/about_us/bayswater.htm>

Milne 2005, Your Home Technical Manual – 3.1 Embodied Energy, Australian
Greenhouse Office, Canberra, Viewed 2 November 2005.
<http://www.greenhouse.gov.au/yourhome/technical/pdf/fs31.pdf>

NAS (National Academy of Sciences) 2005, Global Warming Facts and Our Future,
Marian Koshland Science Museum, viewed 27 March, 2005.
<http://www.koshland-science-museum.org/exhibitgcc/>

NASA 2005, Aerospace Flywheel Development, NASA P&PO, viewed 16 May 2005.
<http://space-power.grc.nasa.gov/ppo/projects/flywheel/techdet.html#advantages>

Nave, R 2005, Ocean Waves, Georgia State Univeristy, viewed 13 August, 2005.
<http://hyperphysics.phy-astr.gsu.edu/hbase/waves/watwav2.html>

OPD 2005, – various titles, Ocean Power Delivery Ltd.


<http://www.oceanpd.com>

OPT 2005, Technology: PowerBuoyTM Operation, Ocean Power Technology Inc.,


viewed 20 September 2005.
<http://www.oceanpowertechnologies.com/technology/>

129
Plummer et al. 2003, Parks Victoria Technical Series: Marine Natural Values Series
Number 1, Parks Victoria, viewed 20 September 2005.
<http://www.parkweb.vic.gov.au/resources/19_1031.pdf>

PMSEIC 2002, Beyond Kyoto – Innovation and Adaptation, Prime Minister’s


Science, Engineering and Innovation Council, viewed 8 September 2005.
<http://www.dest.gov.au/NR/rdonlyres/5AC44381-611C-42D4-8A05-
072DA9E3C8A0/1945/BeyondKyotoreport.pdf>

Pradhan et al. 2002, Large cryogenic storage of hydrogen in carbon nanotubes at low
pressures, Materials Research Society, viewed 22 May, 2005.
<http://www.phys.psu.edu/~crespi/docs/jmr001.pdf>

Prevesic, M 2004, E2I EPRI Assesment – Offshore Wave Energy Conversion


Devices, EPRI (Electric Power Research Institute), viewed 24 May, 2005
<http://www.epri.com/attachments/297213_004_WEC_Device_Assess_Report_Rev1
_MP_6-16-04.pdf>

Purton 2005, MRET Review Submission, Glenelg Shire Council, Portland, viewed 20
September 2005.
<http://www.mretreview.gov.au/pubs/mret-submission230.pdf>

Pyper W 2004, The Aquadam, p.8 in ECOS Magazine, CSIRO publishing, Nov-Dec
edition 2004.

Reid JS & Fandry CB 1994, Wave Climate Measurements in the Southern Ocean,
CSIRO Marine Laboratories – Report 223

RPDC 2005, Energy sources in Tasmania (1998-99), Resource Planning and


Development Commission, viewed 30 October.
<http://www.rpdc.tas.gov.au/soer/image/313/index.php>

130
Sagle A & Freeman B 2004, Fundamentals of Membranes for Water Treatment,
Texas Water Development Board, viewed 24 August 2005.
<http://www.twdb.state.tx.us/Desalination/The%20Future%20of%20Desalination%2
0in%20Texas%20-%20Volume%202/documents/C6.pdf>

Soreneson, B 2000, Renewable Energy – Second Edition, Academic Press.

Sterl et al. 2003, Global Wave Climatology Atlas, KNMI (Koninklijk Nederlands
Meteorologisch Instituut), viewed 20 August 2005,
<http://www.knmi.nl/onderzk/oceano/waves/era40/>

Stuartenergy 2005, Stuart Energy Station: Welcome to the Evolution of Energy, Stuart
Energy, viewed 1 November 2005.
<http://www.stuartenergy.com/pdfs/ses_brochure_20041207_web.pdf>

SWWES (Sun, Wind, Water – Energy Solutions) 2003, Kangaroo Island Energy
Review – Stage 1: Electricity, Infrastructure, Demand, Reliability, KIDB/SENRAC,
Kingscote.

Taylor, N 2003, Eyre Peninsula Water Supply Master Plan, SA Water & Parsons
Brinkerhoff, Adelaide.

Vestas 2004, V120-4.5 MW Offshore leadership, Vestas Wind Systems, Denmark,


viewed 11 August, 2005.
<http://www.vestas.com/pdf/produkter/AktuelleBrochurer/v120/V120%20UK.pdf>

VWS 2005, Energy balances for Vestas wind turbines, Vestas Wind Systems,
Denmark, viewed 11 August 2005.
<http://www.vestas.com/uk/environment/2005_rev/energybalance.asp>

131
Watson, A 2001, The South Australian Wave Rider Buoy and Some Preliminary
Comparisons of Wind and Wave Data, Australian Bureau of Meteorology, viewed 23
May.
<www.dbcp.noaa.gov/dbcp/doc/DBCP-21/DOCS_DBCP21%5C03%20
Andrew%20Watson.doc>

WDE 2005, Wind DeSalter, Wind DeSalter Engineering GmbH, Rendsburg,


Germany, viewed 4 November 2005.
<http://www.winddesalter.net/>

132
Appendix A

Ocean Power Delivery


Pelamis P-750 Brochure

(Intellectual property of Ocean Power Delivery Pty. Ltd.)

133
Appendix B

Statistics of World Energy Use


Taken directly from IEA Key World Statistics 2004

134
135
136
Appendix C

Summary of Wave Energy Information


Resources

Companies and other Non-literature Resources


Bureau of Meteorology - Government
• Wave Buoy Data for Cape du Couedic and Cape Sorrel
• Andrew Watson’s Cape Du Couedic Report, APPENDIX F
• Large amounts of wind data – 100’s of images available for free like the ones
below, that would have needed a lot of data values. The image below was
picked to show the average trend, BOM would be able to provide some
wind/wave climate info that is not provided on website

137
Ocean Prospect
• Ed Mackay Msc Student, Bristol, supplied his wave energy MCP thesis that he
was working on throughout the year for Ocean Prospect.
• Must have some modeling capabilities and connections with OPD

ERA-40 Global Wave Model by the ECMWF


• Built on over 40 years of wind data correlations
• Full 40 year data sets can be purchased for download from
http://data.ecmwf.int/data/index.html
• Global Wave Climatology Atlas (Netherlands) derived from this model:
resolution 0.5 degrees (long/lat), free samples for non-commercial use
available at http://www.knmi.nl/onderzk/oceano/waves/era40
• Poor accuracy closer to the shore
• PLOTS taken from the atlas are reproduced in this report in Chapter 3.1 and
APPENDIX E.

NOAA (National Oceanographic & Atmospheric Advisory) Wavewatch III –


http://polar.ncep.noaa.gov/waves/main_int.html
• Highly reputable source for modeled wave data developed by the US
Department of commerce
• Covering the globe - model available for purchase
• Metocean Engineers have provided full quotes on modeled data (NOAA
WW3) and analysis for specific locations in Southern Australia
www.metoceanengineers.com
• Or modeled WW3 data can be purchased from www.buoyweather.com using
their vitual buoy link

Altimeter Data Retrieval Tool – Laboratory for Satellite Oceanography (LSO)


• Global satellite wave data from http://www.soc.soton.ac.uk/ALTIMETER/
• Apparently free for non-commercial use, have not attempted to register – it is
encoded data (FORTRAN?)

Oceanor World Wave Atlas (WWA)


• High resolution satellite wave energy data verified by buoy data
• Read though pdf file description
• Given trial software – little explored (don’t have satellite data)
• Accuracy is less when closer to shore
• Available for a fee

Ocean Power Delivery (OPD)


• Claim to be a world leader in ocean hydrodynamic modeling
• Have access to Oceanor satellite data

Flinders University Oceanography


• Contact: Dr John Bennett - School of Chemistry, Physics & Earth Sciences,
Position: Lecturer in Oceanography and Meteorology

138
Literature Resources Explored
(Reviewed at different lengths and in addition to exploring ‘literally’ hundreds of
websites)

Duckers, L 2003, ‘Wave Energy’ in Renewable Energy: Power for a Sustainable


Future, Boyle G (ed.), Oxford University Press, Oxford
• Very good resource explaining fundamentals of wave energy and basic
equations
• Has very good outline of different wave energy technologies

Fay JA & Golomb DS 2002, Energy and the Environment, Oxford University Press,
Oxford
• Wave energy chapter is even more basic than Boyle
• Includes 4 empirical formulas relating wind speed to the properties of
generated waves

Johansson, T. et al. (eds) 1993, Renewable Energy: Sources for Fuels and Electricity,
Earthscan, London
• Similar type book to Boyle (Duckers) but with more chapters focussed on
solar (PV especially)
• Has a chapter dedicated to Wave Power – mostly explaining different
technologies and minimal description of wave energy resource

Soreneson, B 2000, Renewable Energy – Second Edition, Academic Press


• Very comprehensive on the origins of renewable energy, lots of figures
• Thoroughly explains wave energy formation and gravity waves
• Has graphs showing fetch-limited energy spectrum of waves from observation
• Only a few pages on ocean waves
• Quite comprehensive description of ocean modeling
• Explains Pneumatic converter and Oscillating Vane Converter (Salter Duck)

- Less extensively reviewed:

Brooke J 2003, Wave Energy Conversion, Elsevier Ocean Engineering Book Series,
Vol 6, Elsevier
• Chapter 2 is a fairly comprehensive easily read explanation of ocean waves
• Appendix 2 contains mathematical descriptions of waves, moderately complex
and mostly applied to deep water
• The rest of the book is foccusssed on Technologies, Pelamis is briefly
included

LeBlond PH & Mysak LA 1978, Waves in the Ocean, Elsevier Oceanography Series
20
• Very heavy on complex mathematical models describing hundreds of
different types ocean wave phenomena and waves types (not just surface)
• Lengthy descriptions of wave-wave interactions and interactions with various
coastal conditions
• Contains hundreds of graphs, some quite interesting

139
• Nothing clear that explains energy levels and utilization

McCormick ME & Kim YC (eds) 1986, Utilization of Ocean Waves – Wave to


Energy Conversion, ASCE, New York
• Session 3 is supposed to contain ‘Wave Climate and Site Considerations’ by
Terry Davis but it says – manuscript not available
• The rest of the book is all case studies of individual developmental
technologies

Count B (ed.) 1980, Power from Sea Waves, Academic Press


• Part II contains 4 chapters explaining the process involved in calculating wave
climate and synthesizing their directional climate
• Contains descriptions of acquiring and analysing buoy data
• Each is based on wind models containing statistally generated wave spectra
resulting from wind fields for deep water only

Evans DV & Falcao AF (eds) 1985, Hydrodynamics of Ocean Wave-Energy


Utilization, Springer Verlag
• Contains a series of case studies into different developmental technologies at
the time – including OWCs, Salter Duck, Tapchans, Power Buoys etc.

Berge H (ed) 1982, Wave Energy Utilization, Proceedings of the 2nd International
Symposium on Wave Energy Utilization, Tapir, Trondheim
• 2 potentially useful studies one called ‘Analysis and description of Wave
Energy Resources’ and the other ‘The Norweigian “Wave Climate Mapping”
Programme
• Both explains trends in wave climatology
• Good use of graphs showing trends from seasons, to daily cycles to minute by
minute variability

List of potentially useful books found on Compendex (copied from unisa library
web gateway):

SAR wave mode data: A new source for ocean and sea ice related
climate research
Lehner, Susanne (Deutsches Zentrum Luft-/R. e.V., Remote Sensing Technology Institute); Konig,
Thomas; Schulz-Stellenfleth, Johannes; Niedermeier, Andreas
Source: Proceedings of SPIE -
The International Society for Optical Engineering, v 5233, 2004, p 126-133
ISSN: 0277-786X CODEN: PSISDG
Conference: Remote Sensing of the Ocean and Sea Ice 2003, Sep 9-12 2003,
Barcelona, Spain Sponsor: SPIE;Sociedad Espanola de Optica, Spain
(SEDO);National Aeronautics and Space Administration, NASA;European Optical
Society (EOS)
Publisher: The International Society for Optical Engineering
Abstract: From raw ERS wavemode data, complex SAR images of size 5 km × 10 km - so
called imagettes - are computed using DLR's BSAR processor. Being collected over oceans
every 200 km along the orbit, imagettes provide a patchy daily global coverage of the world's
oceans. This paper presents DLR's processing chain and results for a three week testing
period covering the detection of ocean slicks and sea ice, the use of cross spectra to derive
wave spectra over open ocean, the detection of one or several ocean wave systems

140
together with its wave length and its propagation direction, and the estimation of single wave
heights. In sea ice covered areas, statistical parameters derived from the SAR are well
correlated to physical measures such as the ice deformation energy and ridge frequency.
Over ice free areas, comparisons between the SAR results and the ECMWF analysis as well
as model results show good agreement in most cases. Observed deviations between SAR
measurements and model results are discussed. As a new application of wave mode data
statistics are presented on the occurence of extreme waves. As soon as ESA will make
available the daily global raw wavemode data for the whole lifetime of ERS-1 and ERS-2, it
will be possible to DLR to derive unique statistics on sea ice and wave parameters for a
period of at least 10 years, which may be even extended by the use of ENVISAT data in
future. Together with near surface wind fields which are also derivable from the imagettes,
this work will be of great benefit for climate research applications. (10 refs.)

DAAT - A new technique for wave directional analysis


Parente, Carlos Eduardo (Ocean Engineering Program, Federal University of Rio de Janeiro) Source:
Proceedings of the International Symposium on Ocean Wave Measurement and
Analysis, v 1, 2001, p 278-287
Conference: Proceedings of the Fourth International Symposium Waves 2001, Sep
2-6 2001, San Francisco, CA Sponsor: Coasts, Oceans, Ports, and Rivers Institute;
Coastal Zone Management Committee; Waves and Wave Forces Committee; Rubble
Mound Structures Committee
Publisher: American Society of Civil Engineers
Abstract: This work presents a new spectral technique for analysis of data from wave
directional sensors. It is well known that one of the main problems in directional analysis is
a poor spatial resolution due to the small number of sensors. The new technique can provide
better resolution by detecting the occurrence of certain events along a record, i.e., waves of a
given frequency range propagating in a well defined direction. The combination of these
events allows for the computation of the energy directional distribution, D(θ), as well as for
the values of spread for different seas and, at last, the directional spectrum. Some examples
of wave climate of the Campos basin off Rio de Janeiro coast using this technique are
shown.

Ocean Wave Measurement and Analysis: Proceedings of the Fourth


International Symposium Waves 2001
Anon Source: Proceedings of the International Symposium on Ocean Wave
Measurement and Analysis, v 1, 2001, 963p
Conference: Proceedings of the Fourth International Symposium Waves 2001, Sep
2-6 2001, San Francisco, CA Sponsor: Coasts, Oceans, Ports, and Rivers Institute;
Coastal Zone Management Committee; Waves and Wave Forces Committee; Rubble
Mound Structures Committee
Publisher: American Society of Civil Engineers
Abstract: The proceedings contains 98 papers. Topics discussed include ocean wave
measurement and analysis, remote sensing, data analysis, theory and statistics, wind wave
generation, wave hindcasting and climate, and wave transformation. (Edited abstract)

Spectral wave climate data from space-borne SAR


Source: Proceedings of the International Conference on
Mastenbroek, C. (ARGOSS)
Offshore Mechanics and Arctic Engineering - OMAE, OMAE98-4452, 1998, 12pp
CODEN: PIOSEB
Conference: Proceedings of the 1998 17th International Conference on Offshore
Mechanics and Arctic Engineering, OMAE, Jul 5-9 1998, Lisbon, Portugal
Publisher: ASME

141
Abstract: A new method is presented that is able to retrieve spectral ocean wave information
from ERS-½ SAR wave mode measurements. The method exploits the fact that a wind vector
obtained from a simultaneous measurement by the ERS-½ scatterometer can be collocated
with each SAR wave mode observation. Combining the scatterometer wind estimate with the
observed SAR spectrum, both the wind sea and swell wave peaks are retrieved. To validate
the method the SAR wave mode observations are collocated with 11 spectral NOAA buoys
located in open water, for a period of over 4 years. It is found that the relative RMS of the
height of waves longer than 200 m is 0.25, with a correlation coefficient of 0.95. Only about
6% of the observed SAR spectra were rejected, in particular in light wind conditions when
non-wave features such as those caused by slicks dominated the imagette. Finally, it is found
that the calibration of SAR spectra acquired at small incidence angle (19.9°) is strongly wind
speed dependent. This is probably caused by a saturation of the sensor in cases of a large
return signal (small incidence angle in combination with high wind speeds). (Author
abstract)

A global view of swell and wind sea climate in the ocean by satellite
altimeter and scatterometer
Chen, Ge (Ocean Remote Sensing Institute, Ocean University of Qingdao); Chapron, Bertrand; Ezraty,
Robert; Vandemark, Douglas
Source: Journal of Atmospheric and Oceanic Technology, v
19, n 11, November, 2002, p 1849-1859
ISSN: 0739-0572 CODEN: JAOTES
Abstract: Numerous case reports and regional studies on swell and wind sea events have
been documented during the past century. The global picture of these common oceanic
phenomena, however, is still incomplete in many aspects. This paper presents a feasibility
study of using collocated wind speed and significant wave height measurements from
simultaneous satellite scatterometer and altimeter sources to observe the spatial and
seasonal pattern of dominant swell and wind wave zones in the world's oceans. Two energy-
related normalized indices are proposed, on the basis of which global statistics of swell/wind
sea probabilities and intensities are obtained. It is found that three well-defined tongue-
shaped zones of swell dominance, termed "swell pools," are located in the eastern tropical
areas of the Pacific, the Atlantic, and the Indian Oceans, respectively. Regions of intensive
wave growth are observed in the northwest Pacific, the northwest Atlantic, the Southern
Ocean, and the Mediterranean Sea. Seasonality is distinct for the climate of both swell and
wind sea, notably the large-scale northward bending of the swell pools in boreal summer, and
the dramatic shift of wave-growing extent from a summer low to an autumn high. The results
of this study may serve as a useful reference for a variety of activities, such as ocean wave
modeling, satellite algorithm validation, coastal engineering, and ship routing, when
information on swell and wind sea conditions is needed. (32 refs.)

142
Appendix D

Descriptive Coastal Wave Photographs

Taken by Dr. Brian Kirke in 1995

Notes:
1. Long swell is produced from greater ocean fetch and distant storms and short
swell is generated from local winds. Both contribute to wave energy but they
also contribute to complex spectrum of height and period.
2. Refraction created by wave particle motion interaction with the sea bed leads
to concentration of waves at headlands and attenuation in bays.

Image A: Long swell waves traveling towards the shore in the top right corner; short
fetch waves traveling toward the top left corner.

143
Image B: Long swell waves traveling toward the bottom left corner – seen as faint
bands in the top of the frame. Short fetch waves traveling to the bottom right. Waves
are refracting around the headland and visibly radiating out in the left side of the
frame.

Image C: Swell traveling straight into the shoreline and strongly refracting down the
channel in the bottom right corner.

144
Appendix E

ERA-40 Wave Atlas Plots


Taken directly from Sterl et al. (2003)

http://www.knmi.nl/onderzk/oceano/waves/era40/

These plots are displayed for the purposes of research only.

The first 4 plots are for global monthly mean Hs and are taken for sample months
representing the middle of Australia’s seasons. The last plot shows the same
information for June only over the South Pacific. It represents the highest resolution
plot for southern Australia although it is so dimensionally skewed that the wave
directions are not very clear.

145
146
147
Appendix F

Extracts from BOM Cape Du Couedic Buoy


Report

Taken directly from Watson (2001)

148
Data Buoy Cooperation Panel : Scientific and Technical Workshop XVII
Perth, Western Australia, 22, 23 October 2001

The South Australian Wave Rider Buoy and


Some Preliminary Comparisons of Wind and Wave Data.

Andrew Watson
Supervising Meteorologist, Bureau of Meteorology, South Australian Regional Office

Abstract
The first wave rider buoy to be operated by the Bureau of Meteorology in South Australian coastal
waters was deployed in November 2000. The buoy is moored to the ocean floor in 80 metres of water, 4
nautical miles west of Cape du Couedic on Kangaroo Island. It is a conventional wave rider, comprising
a spherical stainless steel casing, which houses an accelerometer sensor, process circuitry and a High
Frequency (HF) transmitter. When the buoy moves in response to a passing wave, variations in voltage
are transmitted to a processor which translates them into vertical accelerations and then to wave
heights and periods.

Buoy deployment was a joint project between the Bureau and Lawson and Treloar Pty Ltd, with
contracted assistance from a Kangaroo Island fisherman. The buoy uses HF radio transmission to send
wave data every 26 minutes to a base station at Cape du Couedic. A personal computer at the base
station which logs the transmitted data is polled at regular intervals by staff at the Adelaide Regional
Weather Forecast Centre. This system provides high resolution wave data to Australian weather
forecasters in near real time via the Bureau’s Intranet. Hourly data extracted from the full data set is
provided to the general public through the Internet.

1
Wave data derived from the buoy includes significant wave height (mean height of the highest /3 of
waves in a 26 minute sampling period), maximum wave height, the period of all waves and the period of
the dominant waves in each sample.

All recorded wave data from March to August 2001 inclusive was analysed. Mean significant wave
heights showed little variation from March to July with an average height of 2.6 metres, but increased to
3.4 metres in August. Wave height ranged from a minimum significant height of 0.8 metres to a
maximum of 13.8 metres.

Half hourly wind data from the nearest automatic weather station (Neptune Island, 50 nautical miles
northwest of the buoy) for the same period was correlated with wave data from the buoy. Mean wind
speed (10 minute average) showed little variation from March to July, but was 25% stronger in August
with the average for all months being 16 knots (30 kilometres per hour). Wind speed ranged from calm
to 44 knots, and was positively correlated (R = 0.37) to significant wave height. The strength of the
correlation was dependent on the wind direction. Winds from the westerly (3100 to 2300) and southerly
(2200 to 1400) sectors were generally more strongly correlated to significant wave height (R = 0.45),
although there was notable month to month variation. In contrast, winds from the easterly (1300 to 0500)
to northerly (0400 to 3200) sectors showed almost no correlation to significant wave height (R = 0.02),
with some months actually negatively correlated.

For winds from the westerly and southerly sectors, the average significant wave height (3.2 metres) was
around 1 metre higher than the average height for the other two sectors (2.2 metres). On the
assumption that the height of the locally generated wind waves should be independent of wind direction,
the implication is that winds which have an oceanic fetch are generally associated with a swell wave
component which is not apparent when the winds are of continental origin.

These preliminary results provide some insight into the wave climatology of the area west of Kangaroo
Island, and have implications on wave forecasting for coastal waters of South Australia. Further analysis
of data from the wave rider buoy is planned, with the main aim being to produce a more comprehensive
comparison between wind speed and direction and wave height, for application in operational
forecasting.

149
1. Introduction

The Wave Rider Buoy, manufactured in the Netherlands by Datawell, is recognised as an


international standard sensor for ocean wave measurement. Wave rider buoys are typically used
in coastal waters in depths ranging from 10 to 200 metres. In Australia, wave rider buoys are used
mostly to provide deep water wave data. Buoys are typically moored in a water depth of 80 to 100
metres, between 3 and 6 nautical miles (5 to 12 kilometres) from the shoreline. Several wave
riders are currently operating in Australian coastal waters, mostly along the east coast of the
continent.

The Bureau of Meteorology


has for some time seen a
need for wave data from the
Southern Ocean, immediately
to the south of the Australian
continent. The Observations
and Engineering Branch of
the Bureau approached the
South Australian Region with
an offer to part fund a wave
rider buoy, and assist in its
transport and deployment.
The South Australian Region
then approached interested
external businesses and
agencies seeking the
additional funding required for
the buoy project. Two State
Government agencies agreed
Figure 1 : Wave rider buoy on board deployment vessel
to provide funds toward the
project, enabling a wave rider
unit to be purchased from the private company Lawson and Treloar Proprietry Limited.

Deployment was completed in November 2000 (Figure 1), through a cooperative effort involving
the Bureau’s South Australian Regional Office, Lawson and Treloar and a commercial fisherman
from Kangaroo Island. A fishing vessel owned and operated by the fisherman was used to
transport the buoy and mooring components to the point of deployment.

15S Adelaide

Western Australia
Neptune Island AWS
South Australia

30S
Perth
Adelaide Kangaroo Island

Kangaroo Island Cape du Couedic

Southern Ocean Wave Rider Buoy


36.01S/136.62E 0 50 1 00

kilom etres
105E 120E 135E 150E

Figure 2a : Wave rider buoy location Figure 2b : Wave rider buoy location relative to
relative to Australian continent South Australian coast

150
A suitable deployment point was found at a location 4 nautical miles (8 kilometres) west of Cape
du Couedic, the southwest extremity of Kangaroo Island, South Australia (Figure 2). The buoy and
mooring were deployed in water 80 metres deep. The mooring comprises a length of stainless
steel chain and shackle fixed to the base of the buoy, connected in turn to a rubber shock cord,
poly-line, stainless steel cable and a 300 kilogram anchor weight of stud-link chain. The total
length of the mooring is approximately 150 metres – around twice the water depth. The rubber
shock cord ensures the buoy is free to follow the sea surface as waves travel past. The anchor
weight secures the mooring to the ocean floor.

2. Wave Rider Buoy Description

The “body” of the wave rider buoy comprises a stainless steel spheroid, which houses an
accelerometer sensor, process circuitry and a High Frequency (HF) transmitter. A two metre long
whip antenna is attached to the “top hatch” on the top side of the buoy.

The “heart” of the buoy (Figure 3) is a weighted “platform” which is


suspended by a two axis gimbal (similar to a compass on a ship) in an
electrolytic fluid. This set-up allows the platform to remain horizontal
within the buoy, at least for short period durations typical of sea wave
motion. A flexible bar is attached to the platform. The free end of this
bar attempts to remain in the same position due to its inertia but the
fixed end of the bar moves up and down along with the platform. When
the buoy accelerates in response to a passing wave, the free end of
the bar moves relative to two fixed electrodes and senses variations in
its voltage relative to the electrodes. These variations in voltage are
transmitted to a processor which translates them into vertical
accelerations and ultimately into wave heights and wave periods.

Whilst the buoy system provides detailed information on wave height


characteristics, it cannot provide information on the direction of travel
of waves. It is therefore not possible to discriminate either the direction
of movement of local wind produced waves, or the direction from which
the longer period swell waves come.

3. Communication System
Data transmission is via HF radio (27.7 Mega-Hertz) to a receiving
Figure 3 : Wave rider buoy
station just to the north of Cape du Couedic. Transmissions from the components
buoy are made at 26 minute intervals, with the data from the buoy
recorded by an on-site personal computer at the receiving station. When this computer is polled
by staff at the South Australian Regional Forecast Centre (RFC) in Adelaide, the data are re-
transmitted to a nearby repeater station, which is connected directly into the telecommunications
landline. Data is polled at regular intervals, and once received in the RFC is displayed on
terminals in both graphical and tabular format for use by weather forecasters. The full resolution
graphical wave data is sent to the Internet and available via registered user password access to
the agencies that contributed funds toward the project. Tabular data at approximately hourly
intervals are available to the general public.

151
Appendix G

Description of Wave Data Fields

Supplied by the Bureau of Meteorology

DD/MM/YYYY HH:MN:SS Hs Hrms Hmax Tz Ts Tc THmax EPS T02 Tp Hrms


m m m s s s s - s s m

Following are definitions of data fields and (format) eg.

DD Day of record (xx) 01


MM Month of record (xx) 05
YYYY Year of record (xxxx) 1998
HH hour at start of record burst (xx) 00
MM minute at start of record burst (xx) 01
SS second at start of record burst (xx) 07
Hs sig wave height from the time domain (xx.xx) _5.91
Hrms root mean square wave height from the time domain (xx.xx) _4.24
Hmax maximum wave height in the record - zero upcrossing analysis (xx.xx) _8.88
Tz the zero crossing period from the time domain (xx.xx) 10.61
Ts the period of the significant waves (xx.xx) 13.57
Tc the crest period (xx.xx) _5.51
Thmax the period of the maximum wave (xx.xx) 14.11
EPS spectral width from the time domain (xx.xx) __.85
T02 the period from spectral moments 0 and 2 (xx.xx) 10.00
Tp the period at the peak spectral energy (xx.xx) 15.66
Hrms root mean square wave height calculated from the spectra (xx.xx) _4.43

152
Appendix H

Bathymetric Image of Australia and the


Southern Ocean
Taken directly from the National Geophysical Data Centre (NCDG) website:
<http://www.ngdc.noaa.gov/mgg/image/2minrelief.html>

This image only really indicates the zones around the coast where the sea floor is less
than 3000m depth.

153
Appendix I

Geography of Cape Du Couedic to Cape Sorell

Screen snapshot taken from Google Earth - modified

154
Appendix J

Comparison of Embodied Energy Values


This is a list of embodied energy values collected by a range of studies. The list was created by Joanna
Glover from the following source:

Which is Better? Steel, Concrete or Wood: A Comparison of Assessments On Three Building Materials
In the Housing Sector Department of Chemical Engineering University of Sydney Fourth Year Thesis
By Joanna Glover
Available at the following website viewed 2 Novemeber 2005:
<http://www.boralgreen.shares.green.net.au/research3/chap3.htm>

Each Author has adopted slightly different approaches to determining the embodied energy of steel and
has selected different boundary conditions.

Author Description Embodied Energy


(MJ/kg)
Alcorn Steel, recycled, sections 8.9
Steel, recycled, wire rod 12.5
Steel, virgin, general 32.0
Buchanan Steel, sections 59.0
Steel, rods 34.9
Steel, general 34.9
Steel, pipes 56.9
Lawson Steel 35.0
Basic Oxygen Steel, coated sheet 38.0
Basic Oxygen Steel, stud 38.0
Electric Arc Furnace Steel, reinforcing rod 19.0
Range of Values 8.9-59.0 MJ/kg

Alcorn, J.A & Baird, G 1996, Use of Hybrid Energy Analysis Method for Evaluating the Embodied
Energy of Building Materials; WREC.

Buchanan, AH, Honey G 1992, Environmental impacts of the New Zealand building industry, Dept. of
Civil Engineering, University of Canterbury, Christchurch, N.Z.

Lawson, B 1996, Building materials, energy and the environment : towards ecologically sustainable
development, Royal Australian Institute of Architects, Red Hill, A.C.T.

The following table contains a comparison by Lawson (1996) of the energy required
for the transportation of steel for the countries Canada and UK. It is assumed that the
energy values quoted consist mostly of the calorific content of fuel used (diesel). It is
assumed that the values for Canada would be similar to those for Australia.

Energy consumption (MJ/tonne.km)


Form of Transportation Canada UK
Road 1.18 4.5
Rail 0.49 0.6
Ship 0.12 0.25

155
Appendix K

Environmental Impact Assessment - Generic


Scoping Study

156
Appendix L

Industrial Experience Report

Onesteel’s Whyalla Steelworks – Summer 2004/5

Note: An effort has been taken not to include any confidential or commercially
sensitive information in this report.

My 3 months of industrial experience was undertaken with Onesteel at the Whyalla


Intergrated Steelworks at the end of my third year of study. During these 3 months I
lived in the UniSA accommodation at the Whyalla Campus. I worked at the Rolling
Mills Technology and Development department under a supervisor and a mentor.
There were 11 ‘vacation student engineers’ from all over Australia (one from New
Zealand), and one other (mech eng) at the Rolling Mills in a different department. The
bulk of my time and effort went to two separate design projects and drafting these on
CAD. Overall I had a very diverse range of experiences and a thorough introduction
to the operations of a major steelworks. To begin with I’ll provide a background of
where I worked and then more detail about my design projects.

The rural city of Whyalla has its economic output is centered on Onesteels operations.
Onesteel is an independent company born out of the Broken Hill Proprietary
Company Limited (BHP) in 2000. The history of Whyalla began with the former BHP
mining iron ore from the middleback ranges 50kms to the west over 100 years ago. In
1940 BHP expanded its industry by building a blast furnace to produce iron and a
shipyard, (which closed in 1978) and then the steelworks started production in 1965.
The name Onesteel was coined as to express the collaboration of the various
businesses under its name. It has operations in over 170 locations in Australia and
New Zealand and around 7,000 employees. However the Whyalla Steelworks is the
engine room for Onesteel’s operations. It is located on a large area of land (~1000
hectares) north of the town on the coast of the Eyre Peninsula. The annual steel output
at Whyalla is 1.2 million tonnes, making Whyalla the second largest steel producer in
Australia after Bluescope’s Port Kembla operations in NSW (also formerly BHP).

Now I will provide a brief overview of the operations of the Whyalla steelworks, to
put the Rolling Mills in perspective. Iron ore mined in the middleback ranges is
continuously feed by train to Ore Processing at the base of the steelworks. The
hematite ore is pelletized at the Pellet Plant (or Pink Palace) so that it can be used as
feed for the Blast Furnace. Coal is cooked in a series of ovens to produce coke at the
Coke Making plant. The pellets, coke and also limestone are fed to the Blast Furnace
where molten iron and slag is produced. The molten iron is poured from the base of
the furnace into Treadwell ladles and sent to Steelmaking by rail. Steel is then
produced by the Basic Oxygen Steelmaking process and sent to the Casters. Cast slab
and bloom is sent to the Rolling Mills (explained later) and cast billets (smaller) are

157
mostly sent to the Mills (rod, bar and wire) in Newcastle. Most of the operations at the
steelworks occur 24 hours and 7 days a week, requiring continuous shift work labor.

Before I began work I was given a 3 hour general safety induction followed by a drug
screening and a brief tour of the steelworks site (only 2 hours). Safety is at the centre
of Onesteel’s business ethics with a very strong focus on preventing injury. The safety
training provided an overview of the workplace safety standards and practices over
the whole business. Many aspects of which would apply to most industrial
workplaces, but also many important pieces of unique information were provided for
safety when working at various parts of the steelworks. Whyalla’s steelworks is a
heavy industry and many plants onsite present hazards to human health. Working in
potentially extremely dangerous conditions requires special procedures, training and
often a permit or strictly licensed personnel. I was introduced to the safety lockout
system, using coloured identification padlocks, which I was to use a number of times.
Location boards, identification tags, emergency procedures & gather points were also
covered. The safety record at the steelworks is at an all time best since Onesteels
ownership began. Targets for LTI (lost time injury) are repeatedly exceeded and the
death toll is far lower than that of decades past, and dramatically less than the horrific
numbers before the 60’s. It did not take long to realise how important the workplace
safety training was. Every day just to get to the office I needed to wear PPE (hard hat,
safety glasses, earplugs, reflective clothing & steel capped boots) while crossing
through major sections of the plant, while constantly keeping an eye on crane hooks
and industrial vehicle traffic.

On my second day I was given an induction to the whole Rolling Mills area which
provided me with a solid overview of the plant operations, which I will briefly
describe. Slabs and blooms from the caster lie in outdoor stockpiles and are crane fed
into the Reheat Furnace. This furnace, usually fuelled by coke ovens gas, heats the
steel to about 1260°C where the steel becomes soft and ‘plastic’. The steel is then
transferred into the Mill (a large & long building) where it is transported by rollers
into a series of 4 mill ‘Stands’. Each stand supports a set of rolls which shape the bar
into the desired section by repeated passes in both directions. It takes a few minutes
before the bar leaves the last stand (Finishing Mill) at the final cross-sectional
dimensions and even though it has cooled considerably, they are normally still
glowing red. The bar leaves the mill building and moves outdoors to the Hot Saw
plant. Large circular blades cut the hot steel to rough lengths and then they are
transferred to the cooling beds. Once cooled, depending on the section, the bars are
run through straighteners, hardeners, inspections, cold saws (to final lengths) and then
to the stacking beds ready for shipping. The Rolling Mills produces the following
steel sections in a range of sizes ready for the customer: Rail & Sleepers, Universal
Beams, Universal Columns, Parallel Flange Channels, Angles (equal and unequal).
The Rolling Mills is a mixture of machinery and structures from the 60’s, when it was
built, through to new and advanced machinery with computerized and automated
control. The number of employees at the Rolling Mills and the rest of the steelworks
have dropped dramatically over the last few decades with the introduction of
automated machinery.

During the first week I was introduced to my supervisor, mentor (department


manager) and a number of the other employees of the Technology and Development
department. My supervisor, the Principal Roll Designer, gave me a detailed tour of the
Mill from #1 to #4 Stand and the operators within the pulpits (control rooms). Then he

158
allowed me some time to become familiarized with the CAD package IDEAS Master
Series 3 (’96 on Unix) that I used primarily, but I also used AutoCAD (already
familiar with). The offices were a mix of Windows PC’s, Unix plant monitoring and
control systems and CAD stations (I moved from computer to computer). I was taken
on much more detailed tours of the rest of the Rolling Mills by other workers in my
office for the remainder of the week. I was separately inducted to the Reheat Furnace
Area and also taken to the Mill Service Shop (some drive from the Mill) where giant
CNC lathes profile the rollers as they wear. In the process I was given a thorough
description of these employees and introduced to a number of other workers
throughout the mill. I had the opportunity to ask hundreds of questions and observe a
staggering array of machinery in operation.

As stated in the introduction, the bulk of my time and effort went to two separate
design projects. My supervisor was a very busy person, with a big whiteboard packed
with an overwhelming amount of tasks. These two projects were set aside for me
months before I began work as they required far too much time from my supervisor.
The first was called ‘Bottom Flange Cooling Design for Universals’ which I termed
BFC and the other was called ‘Tilting Apron Roller Height Adjustment Mechanism’
or TRA. I worked on these two projects concurrently right through to the last week.
For both of these projects I needed to inspect certain ‘roll changes’ that were rare or
happened at night. I often needed to put work on hold until I could make these
important measurements which required a full safety lockout during scheduled
downtime. I created two detailed PowerPoint documents reporting these projects and
presented them to the Technology and Development Department and others (a total of
about 25 people, many were engineers). I was given very positive feedback. The next
two paragraphs will describe each project and the process that I went about the work
and what I produced.

The BFC project was my top priority work and was the most experimental, with
nothing performing the same function existing anywhere in the steelworks. All of the
hours on this work were documented under research and development. It was
conceived by my supervisor on his work trip to Japan’s Nippon Steelworks the
previous year (where they have something like it operating). In short the goal of the
project was to produce a design that would allow better control over the dimensions of
Universal Sections (beams and columns). As the bar moves through the mill the top
flanges are cooled more than the bottom creating a number of difficulties in
controlling bar behavior further down the mill. Although it will be an expensive
project to implement it has the potential for huge financial savings by eliminating lost
product (scrap and crop length) and was given quite some importance within the
department. I began by eliminating major design alternatives to the one suggested by
consultation with my supervisor. The design called for a set of water sprays aimed at
the bottom flange of each of the whole range of different universal sections. There
were a number of overriding space constraints (which I identified first) and also the
whole design had to be integrated with setup procedures and made ‘user friendly’ and
‘fool proof’ for operators. Also it needed to be robust, reliable, semi-adjustable and as
cheap as possible with minimal maintenance. After exploring a number of different
solutions including a single elaborate fully adjustable mechanism, I came up with
three separate arrangements to cover certain groups of sections. These arrangements
are then ‘plugged and pinned’ into existing equipment that needed modification. I had
a number of consultations with my supervisor and other relevant engineers in Rolling
Mills for advice and feedback on my work. I had to search over dozens of drawings

159
(electronic and hard copy) and make many ‘on site’ measurements as often the
drawings weren’t available, modified or incomplete. My final design for each
arrangement where a pair of spray header pipes each fitted with a set of 36 nozzles
carefully angled to spray water over the sections required. After several revisions my
design refined to the most basic yet functional form for minimal cost. Considerable
time was spent also sourcing the water and piping routes and what kind of work
would be needed to integrate the design. Some of the calculations made for this
project were pumping head loss, spray trajectory and surface flux to maximise heat
transfer.

The TRA project left more room for creativity although it was of less importance than
the BFC. There were more alternative solutions and the whole design was conceived
from first principles. At the exit side of the last mill stand (Finishing Mill) the first
coasting roller is suspended in a ‘tilting apron’. This could be described as a platform
that can be pivoted upwards (much like a small opening bridge) to allow space for the
Universal Mill housing (about 50 tonnes) to be lifted in and out by crane every time a
section change to universals is made. One side of the apron is a clevis connected to a
reversible hydraulic piston for raising and lowering the apron. The first calculations
were made to determine the loading on this piston and the moment at the pivot. This
was done by finding the distribution of component weights on the apron. Now, the
only means in place for fine adjustment for the height of the roller at the end of the
apron was very crude. It involved shoving small steel blocks of different thicknesses
under the frame of the apron before it was lowered. This was unacceptable and was
disrupting dimensions of the final product and was causing roll damage and
occasionally leading to scrapped bars. In addition to this the whole apron, installed in
the 60’s, is heavily rusted; riddled with cracks and looked as though it may fall apart.
The design required an adjustable mechanism built into the apron which could raise
and lower it while the Mill was in operation. After exploring a wide range of possible
solutions, an optimised design was identified and pursued. At the heart of the design
is a 2 ton worm screw actuator (jack) which is to be operated manually by a cranking
handle. The actuator is not designed to take high loads over such a small section of
thread with dynamic shock. So a sealed and greased ‘wedge-plate’ system was
designed which adjusts the height of the platform via a pair of mating curved bearing
plates mounted beneath the apron. This arrangement is designed take very high loads
(over 10 tonnes) and makes much better use of the actuator travel and prolongs its life
by effectively eliminating static loading. The arrangement is constructed using a range
of shaped plates all welded together and then welded to the apron’s foundations with
additional webs for strength. Significant structural reinforcements were also
incorporated into the design with the aim of improving the performance and life of the
whole tilting apron. The final design was well approved of by the principle
mechanical engineer, who I had several consultations with throughout the project.

In addition to two detailed PowerPoint presentations, I produced 4 large CAD


drawings for these projects along with dozens of sketches with calculations. The
drawings showed full assembly details with a bill of materials and each was entered
into the Onesteel records. I accumulated over 200 hours on the CAD stations alone.
Earlier on in the experience I spent over a week working on the statistical analysis of
thousands of Rail dimensional samples using Excel. Most of it I undertook through
my own initiative and incorporated it into existing macros for proactive reports. The
histograms I produced were used by my supervisor to identify a long term problem
and some direction for solutions. I was provided with some training using Visual

160
Basic macro programming which has since become valuable in my final year project
at UniSA. I attended and participated in monthly network meetings of which safety,
productivity and incident reports were on the main agenda. Over the 3 months I had to
communicate with a range of employees working in roles covering a wide spectrum.
At the end I was asked to produce a general feedback report on my experience for the
attention of the Manager, Steel Products. As encouraged by the university I kept a
daily log or diary containing a brief record of my daily activities (can be provided on
request).

Nearing the end of the experience all of the vacation students were taken on a series
of detailed plant tours over two full days. During the first day tour I was taken to the
ore mine at Iron Duke, then to the Pellet Plant and then Coke Making. On the second
tour I went to the Blast Furnace, then to Steelmaking and Casters and then back to the
Rolling Mills. The students who were working at each plant were encouraged to
participate in delivering the tours. Looking back this was one of the most valuable
experiences of all and was critical to my understanding of how the whole steelworks
functions. The sheer scale and energy of a lot of the machinery, equipment and
structures was eye opening to say the least. Overall my experience at Onesteel’s
Whyalla Steelworks was a very positive and rewarding one. It has provided a major
contribution to my professional and personal development. I learnt a great deal from
the experience in a short time that I could never have obtained from tertiary studies
alone. From my interactions with students who obtained experience elsewhere, I have
the impression that Onesteel’s Vacation Program was of very high quality. I would
recommend it to any 3rd year mechanical engineering student.

On the Following Page is a performance summary written by my mentor.

161