Offshore Wind Resource Assessment Off The South African Coastline

International Journal of Mechanical Engineering and Technology (IJMET)
Volume 10, Issue 06, June 2019, pp. 95-119 . Article ID: IJMET_10_06_006
Available online at http://www.iaeme.com/ijmet/issues.asp?JType=IJMET&VType=10&IType=6
ISSN Print: 0976-6340 and ISSN Online: 0976-6359
© IAEME Publication
OFFSHORE WIND RESOURCE ASSESSMENT

OFF THE SOUTH AFRICAN COASTLINE
Freddie L. Inambao and Kumaresan Cunden
School of Mechanical Engineering, University of Kwa-Zulu Natal
Durban, South Africa
ABSTRACT
The world is undergoing a paradigm shift as more people are becoming aware of
energy consumption patterns, reinforcing the need for developing cleaner and more
sustainable ways to generate electrical energy. Globally, the development of onshore
wind farms is sometimes impeded by factors such as aesthetic impact, acceptance by
the public, the threats to surrounding biodiversity, noise from the power plant and
possible land use conflicts. Due to these concerns, offshore wind plants have been
developed. Offshore wind energy is generally greater in comparison to that of onshore
wind energy because the wind speeds offshore are generally higher and more constant
with fewer obstructions to the wind resource. The offshore wind potential for South
African coastal regions was investigated and analysed in this study. Various factors
such as shipping routes, oil and gas exploration fields and possible transmission
connection points were taken into consideration before selecting four data collection
sites. The predominant wind direction, mean wind speed, wind shear and spatial
geographic information was analysed for each site. The sites’ wind direction did not
have any similarities, with each site having its own prevailing wind directions. Within
the 50 m hub height, Site 2 showed the best potential based on the power density. Site 1
and Site 3 showed similar power densities to each other with Site 4 showing the lowest
power density. The distance to shore ranged from 200 km to 500 km with a steep
continental shelf drop to a depth of approximately 3 000 m. The study conducted shows
that there is offshore wind potential off the coast of South Africa. Energy generated by
this method could assist South Africa to increase access to energy, reduce expensive
transmission line losses to coastal provinces, and assist the country to transition
towards a more sustainable future energy mix in line with developed nations.
Keywords: Offshore wind, Resource Assessment, WAsP.
Cite this Article: Freddie L. Inambao and Kumaresan Cunden, Offshore Wind
Resource Assessment Off the South African Coastline, International Journal of
Mechanical Engineering and Technology, 10(6), 2019, pp. 95-119.
http://www.iaeme.com/IJMET/issues.asp?JType=IJMET&VType=10&IType=6
http://www.iaeme.com/IJMET/index.asp 95 [email protected]
1. INTRODUCTION
The world is undergoing a paradigm shift as more people are becoming aware of energy
consumption patterns, reinforcing the need for cleaner ways to generate electrical energy. The
International Renewable Energy Agency (IRENA) is an intergovernmental organisation which
supports countries in their efforts to attain a more sustainable energy future. Figure 1 was
constructed using IRENA’s Renewable Energy Statistics databases [1]. The trend in Figure 1
indicates rapid growth in the installed renewable energy capacity over the past decade across
all renewable resources.
Figure 1 Total global renewable energy installed capacity

During the first five years of the millennium the European market dominated the renewable
energy portfolio. However, the Asian and the North American markets have now gained
significant market share. Figure 2 depicts the evolution of the renewable energy capacity market
share across the major continents of the world. Asia has now taken the lead with the most
installed renewable capacity comprising 39.85 % (204.6 GW) of the market in comparison to
the Europeans with 33.2 % (170.6 GW) [1].
Figure 2 Market share of renewable capacity 2000-2016

Onshore wind and solar photovoltaic (PV) technologies have experienced the largest
amount of growth over the last decade in relation to other renewable technologies, as seen in
Figure 3 [1]. This can be attributed to the maturity of technologies and rapid advancements in
the respective fields of study, as well as the ease of deployment in comparison to larger thermal
plants such as bioenergy and geothermal power plants.
Offshore Wind Resource Assessment Off the South African Coastline
Figure 3 Renewable energy technology growth

In many regions of the world, the development of onshore wind farms are sometimes
impeded by factors such as visual impact, acceptance by the public, threats to the surrounding
biodiversity, noise from the power plant and possible land use conflicts [2]. The multiple
permutations of these conflicts are likely to hinder the future development of onshore wind
farms. Due to these possible future impacts, as well as the added benefit of increased specific
production (1) (kWh/kWP), major developments in the world are moving towards offshore wind
farms [3, 4].
The potential of offshore wind energy is greater than that of onshore wind energy. This is
because the wind speeds found offshore are generally higher and more constant with less
obstructions to the wind resource. Added to this, there are the fewer concerns regarding noise
pollution, aesthetic imact and most other types of land-based turbine restrictions [5]. The
development of offshore windfarms has led to an increase in turbine swept areas from larger
turbine blades resulting in an increase in generator capacity harvesting more energy per square
meter. Though the advantages are numerous, some of the disadvantages of offshore wind
energy development include expensive marine foundations and the high costs of onshore
electrical grid integration [3]. For the purpose of comparison, Figure 4 shows the contribution
of energy produced by both areas of supply, normalised by the installed capacity during that
period.
Europe contributes a large amount to the growth of offshore wind installations because of
limited land area available for onshore developments [6], [7]. The largest offshore wind farm
in the United Kingdom was commissioned in 2012 and is the London Array, consisting of a
total of 630 MW of generation capacity. A further 370 MW is planned for phase 2 development
[8]. China’s first commercial offshore wind project is located close to the Donghai Bridge in
Shanghai totalling 102 MW in capacity with grid supply since June 2010 [9], [10].
Figure 4 (a) Installed capacity; (b) Normalised production

This paper aims to evaluate offshore wind energy potential for the development of a
conceptual offshore wind energy power plant facility off the South African coast. The
investigation aims to understand the average wind speed, dominant wind direction and wind
shear effects of particular locations in the region. Shipping routes and relevant bathymetry will
be investigated to better understand the environmental conditions which need to be overcome
before development of an offshore wind farm.
2. LITERATURE REVIEW
Before developing the methodology for the offshore wind resource evaluation, an
understanding of best practice was required. Murthy and Rahi [11] conducted a review of
offshore wind energy assessments regarding the main characteristics required for a resource
assessment. Their work outlined the various methodologies used for wind power projects and
the uncertainties associated with wind energy assessments. The paper gave a basic
understanding of wind behaviour through periods of change (diurnal, seasonal, monthly and
annual). The authors state that a minimum data log of 1 meteorological year is required for an
assessment, however, more data would lead to a more accurate estimation of energy potential
for the given region.
Sharma and Ahmed [12] conducted research on the wind energy potential for the Fiji islands
of Kadavu and the Suva Peninsula. The authors gathered mean wind speeds and predominant
wind directions for each site based on 18 month and 12 month investigation periods
respectively. Wind shear effects were also investigated to understand the variation of shear
force with respect to height. The authors utilised the WAsP simulation tool to simulate a high
resolution (5 km2) resource map for both regions. The result of the simulations showed a good
potential for wind energy production.
Lima et al. [13] sought to estimate the offshore wind energy potential for Ceara in Brazil in
an effort to increase the maturity of the offshore wind energy sector. The study utilised the
Regional Atmospheric Modeling System (RAMS) to estimate the average wind speed, direction
and power density for the area. The study considered the bathymetry and the shipping traffic
for the Ceara region. The authors evaluated the average wind speed for three periods consisting
of El Nina, El Ninõ and Neutral years, each of which were evaluated through seasonal changes.
The results showed an average wind speed of 8 m/s and a power density of roughly 720 W/m2
no matter what the period.
Werapun et al. [5] conducted an offshore energy potential study for southern Thailand. The
study followed a similar methodology to the previous authors’ utilising a 120m high
meteorological mast measuring data at different heights along the mast. The average wind speed
was seemingly low (4.28 m/s) with the dominant wind direction stemming from the North
which resulted in wind power density being 85 W/m2. The area was simulated using the WAsP
simulation tool for nine base cases for wind farm layout and the authors found that the capacity
factors of the simulation ranged from 0.98 % to 2.68 %.
Kim et al. [14] investigated the potential for offshore wind farm site selection aimed at
finding the feasibility of an offshore wind farm site around the coastal regions of the Jeju
islands, South Korea. The evaluation categories in this study were: energy production and
economics, protected areas, human marine activities, and the marine ecology of the area. The
researchers concluded that the number of feasible areas for offshore deployment was low when
utilising all factors of all the categories in comparison to just using the energy potential and
economics of the region alone.
Mahdy and Bahaj [15] identified that there is a global gap regarding the assessment of
offshore wind potential sites and thus proposed a new methodology of assessment for potential
offshore sites. The methodology was based on the analytical hierarchy process in conjunction
with spatial assessment within a GIS domain. The methodology was developed with the aim of
assisting the scaling up of renewable energy from 1 GW to 7.5 GW in Egypt by 2020. The
authors hypothesised that the increase in renewable energy would come from larger offshore
wind installed capacity. Areas identified were potential sites around the Red Sea which was
duly estimated to be able to accommodate 33 GW of installed capacity. The researchers
concluded that the methodology which was developed could be applicable globally to produce
adequate offshore wind suitability maps for potential wind power locations.
3. METHODOLOGY
The main aim of this study was to identify an ideal offshore location to situate a large floating
offshore wind farm to supply coastal regions of South Africa with clean renewable energy. The
conceptual assessment of this task was conducted based on methodologies found in the
literature. The methodology was governed by three criteria for site selection: impact on shipping
routes, spatial proximity to the electrical grid, and possible impact on future offshore oil and
gas exploration.
Meteorological data set was obtained from the Global Wind Atlas which was developed by
the Technical University of Denmark, Department of Wind Energy [16], in WAsP data file
format. The data set consisted of 624 individual meteorological points containing wind resource
data (Figure 5). The wind resource data was interpreted using WAsP simulation software. The
data set contained wind resource data for 12 sectored wind roses at each location and associated
wind profile frequency distributions at 3 hub heights of 0 m, 50 m and 200 m respectively.
The analysis of the wind vector data was first filtered by understanding the prevalent wind
direction through evaluating the wind rose generated for the dataset. The wind rose along the
eastern coastline of the country exhibited two distinct wind directions – either stemming from
the north north east or from the south south west.
Figure 5 Meteorological data points

Once the resource data was obtained, the shipping routes were overlaid to gauge high and
low-density shipping paths (Figure 6). The potential sites were located in positions where there
was little to no marine traffic. The shipping routes were constructed using images obtained from
the Marine Traffic website [17]. It is evident that the South African coast line experiences a
high volume of marine traffic.
Figure 6 Shipping routes [17]

The second criterion was spatial proximity to the national electrical grid. The transmission
electrical grid was sourced from the Africa Electrical Transmission Network (AICD) and
implemented in Google Earth as shown in Figure 7 [18]. The coastal ports of South Africa are
located at points a, b and c which represent the KwaZulu-Natal, Eastern Cape and Western
Cape provinces respectively [19].

Figure 7 Transmission electrical grid South Africa [18]

From the Eskom transmission expansion plan 2016 to 2025 [19] it was found that point
KwaZulu-Natal, Eastern Cape and Western Cape provinces in Figure 7 would increase by 841
MW, 609 MW and 656 MW respectively over the 10-year period. As the bulk of the power
stations supplying the country are located inland, the transmission losses incurred while
transmitting energy to coastal regions are high. It was hypothesised that an offshore wind farm
would be able to minimise transmission losses by supplying energy to coastal regions. Added
to this was the resultant decrease in the loading on the onshore electrical grid. This may allow
for future capital to be transferred from many smaller distribution scale grid upgrades to
prioritised transmission upgrades so as to address unserved customers or planned grid
expansion.
The third criterion was the location of offshore oil and gas fields. Although countries are
reforming to move away from fossil fuels, the commodity trading business around oil and gas
of any particular country has significant benefit to that country’s economy. Potential sites were
identified in areas where there would be minimal impact on exploration of the area by interested
organisations / companies. Figure 8 depicts the offshore oil and gas identified sites within the
coastal region of South Africa [20].
Figure 8 Offshore oil and gas – identified sites [20]

Potential sites for offshore wind farms were assessed based on all the above-mentioned
criteria. Figure 9 shows the identified potential offshore sites. The meteorological points have
a spatial resolution of roughly 50 km x 60 km. Site 1 and 2 slightly overlap with two of the
identified oil and gas areas, however, this was only due to the nature of the resolution of
meteorological data.
Figure 9 Identified potential sites

Once each site’s simulated wind data was analysed from each of the meteorological points
within the selected boundaries, an average wind speed for hub heights 50 m, 100 m and 200 m
was calculated. Wind speed is a key factor in site analysis as it is indicative of the potential
within the region for energy extraction.
4. SITE INVESTIGATION & ANALYSIS

This section of the investigation analyses each of the potential sites which were assumed based
on the above methodology and depicted in Figure 9 above. Wind resource data was difficult to
acquire. The use of satellite imagery tools (ASCAT, QuickSCAT, RapidSCAT etc.) did not
contain relevant data which could be readily used for the analysis of the wind potential at the
chosen sites. The International Renewable Energy Agency (IRENA) data was more suitable.
Irena is a free, online portal for the high-level assessment of renewable energy resource
potential at a chosen location / site [21].
The wind resource data was obtained from the Technical University of Denmark’s (DTU)
Global Wind Atlas portal which uses data from IRENA [16]. Data from this source was also
used to develop the Wind Atlas of South Africa – WASA [16] which was developed for onshore
wind analysis of the South African coastline by various South African research bodies such as
the CSIR, South African Weather Services, University of Cape Town and the DTU Department
of Wind Energy. The wind atlas was constructed using various onshore meteorological sites in
conjunction with conventional forecasting tools such as the WAsP wind simulator.
The online portal allows for the rough selection of points for data collection. The potential
sites were mapped as estimated polygon shapes based on the co-ordinates of each of the
boundary points. The resultant wind statistical data was downloaded in WAsP format to be
analysed further. Wind speed and wind direction were analysed for each of the sites depicted in
Figure 9.

4.1. Wind Direction
4.1.1. Site 1
Site 1 is located off the eastern coast of Durban, roughly 200 km offshore from the Durban
harbour in a south east direction or 115º bearing. The site has an estimated perimeter of 600 km
and an estimated surface area of ± 21 000 km2. Figure 10 is a graphical representation of Site
1. The site encompassed 11 of the meteorological data points obtained from DTU Global Wind
Atlas [16].
Figure 10 Potential – Site 1

A wind rose is a vector plot which denotes direction and frequency respectively. A wind
rose developed from the wind resource data comprises 12 sectors where Sector 1 encompasses
a bearing of 345º to 15º and thus the midpoint for Sector 1 of the wind rose plots was 0º with a
15º arc span on either side (Figure 11). Once a common understanding of the demarcation of
the sector boundaries was developed, the sectors within the wind rose were utilised as means
of reference to establish which direction the wind stemmed from for all the potential sites under
investigation.
Figure 11 Site 1 – Wind rose plots

The most predominant wind directions were found to be Sectors 2 and 8 of each of the wind
roses for Site 1. The average wind speed for the area was calculated based on the mean wind
speeds of each of the meteorological stations analysed. The average wind speed for hub heights
50 m, 100 m and 200 m were 9.19 m/s, 9.25 m/s and 9.34 m/s respectively.
4.1.2. Site 2
Site 2 is a larger site than Site 1. It is located roughly 300 km south of Port Elizabeth. It has an
estimated perimeter reading of 1 250 km and an estimated surface area of ± 79 000 km2. The
site is located in a potential retroflection zone (2) for ocean currents as found by Cunden [22].
Figure 12 is a graphic representation of the potential of Site 2. As seen in Figure 9, Site 2 has
minimum shipping route impacts but may lie on a possible potential oil and gas field. From the
meteorological data collected, the site comprised 36 individual meteorological data points
which were analysed at various hub heights.

Within Site 2, the prevailing wind direction was found to be in Sectors 3, 4, 9, 10, and 11
(Figure 13). This indicates that the wind direction changes through an annual cycle. The average
wind speed for hub heights 50 m, 100 m and 200 m were 9.47 m/s, 9.55 m/s and 9.68 m/s
respectively.
Figure 13 Site 2 – Wind rose plots

4.1.3. Site 3
Site 3 is located roughly 290 km south (± 188º bearing) of Cape Town (Figure 14). The site has
an estimated perimeter of 463 km and an estimated surface area of ± 13 332 km2. The site is
the smallest of the four identified potential wind farm sites.

At Site 3 the prevailing wind directions were in Sectors 5 to 11, with the highest occurring
in Sectors 9, 10 and 11 (Figure 15). The distribution of wind over a span of 180º may be
attributed to the convergence of two significant climate zones, namely, the Atlantic and Indian
ocean zones, mixing in South Africa’s southernmost region. The average wind speed for hub
heights 50 m, 100 m and 200 m were 9.12 m/s, 9.12 m/s and 9.27 m/s respectively.
Figure 15 Site 3 – wind rose plots
4.1.4. Site 4
Site 4 is located an estimated 570 km north west (± 298º bearing) of Saldanha Bay and is the
furthest potential site out of the four. The site has an estimated perimeter of 807 km and an
estimated surface area of ± 37 419 km2. The site has a minimum impact on oil and gas
exploration sites and shipping routes as seen in Figure 16.


Site 4’s wind rose plot is depicted in Figure 17 where it can be seen that the predominant
wind direction is south easterly (Sectors 6 and 7). This may be a result of the mixing climatic
conditions further south leading to a retroflected mass of wind moving towards the north from
the Cape peninsula region.
Figure 17 Site 4 - wind rose plots
4.2. Wind Speed

To gauge the wind speed potential in each of the site areas, each of the meteorological points
were mapped for hub heights of 50 m, 100 m and 200 m and an average wind speed was
calculated for each hub height respectively. It is known that wind shear profile has a logarithmic
nature and may be approximated in mathematics by means of the log wind profile equation
which takes into account factors such as surface roughness and atmospheric stability. The
equation which governs the logarithmic nature of the wind profile is as follows:
𝑢∗ 𝑧 (1)
𝑈= [ln ( ) − 𝜓(𝜁)]
𝑘 𝑧0
Where:
𝑈 Mean wind speed (m/s)
𝑧 Height at which mean wind speed is located (m)

𝑢∗ Frictional velocity (m/s)

𝑧0 Surface roughness (-)
𝜓(𝜁) Integrated stability function governing momentum (-)
The integrated stability function comprises the Obukhov Length Scale which includes
further parameters such as sea surface temperature, kinematic heat flux over the surface body
as well as the cube of the frictional velocity [24]. However, these factors are not easily found
in literature or readily available and for the analysis being conducted a more simplified version
of the above was utilised in order to obtain a plausible understanding of the wind shear profile
for the site areas.
To extrapolate the wind shear profile the more common Wind Power Law was used [25]:
𝑧 𝛼 (2)
𝑈 = 𝑈𝑅 ( )
𝑧𝑅
Where:
𝑈 Mean wind speed (m/s)
𝑈𝑅 Reference wins speed (m/s)
𝑧 Height at the mean wind speed (m)
𝑧𝑅 Reference height (m)
𝛼 Empirical Coefficient accounting for the atmospheric stability (-)
Equation 2 has been found to be suitable to extrapolate wind shear for hub heights less than
100 m [23]. The 50 m hub height measurement was used as the first basis for extrapolation for
each of the sites. An iterative solver was used to extract the atmospheric coefficient required.
Figure 18 depicts a possible representation of the different wind shear profiles for each of
the sites. The extrapolation shows the changing wind shear at various hub heights below 100 m
based on the arithmetic average of mean wind speeds located within the boundaries of each site.
As expected, the mean wind speed located offshore is significantly higher that onshore with a
gradual increase in wind shear gradient.
Figure 18 Extrapolated wind shear profiles

Wind resource is relatively unpredictable and erratic by nature; however, on consistent
monitoring and investigation of terrestrial based meteorological stations, a best fit curve was
found to adequately describe the wind resource which is based on the 2-parameter Weibull

distribution. As shown in literature, the 2-parameter Weibull statistical distribution is the most
common way of understanding wind potential [26−28].
The Weibull distribution gives one an appreciation of the probability of wind speed
occurrences at a chosen site. The function requires three parameter inputs, one being a variable
and the two other parameters being static which define the distribution. The Weibull distribution
is defined as follows:
𝑘 𝑈 𝑘−1 −( 𝑈 )𝑘−1 (3)
𝑃(𝑈, 𝑘, 𝑐) = ( ) 𝑒 𝑐
𝑐 𝑐
Where:
𝑈 Wind speed (m/s)
𝑘 Shape parameter (-)
𝑐 Scale parameter (m/s)
The shape parameter of the distribution defines how the probabilities of wind speed are
spread throughout the distribution. It is generally found that the larger the shape parameter the
closer the distribution tends towards becoming Gaussian in shape. The shape parameter for
wind distributions is usually near 2 [29]. The scale parameter defines the most probable speed
in the distribution (the mode value). Traditionally with regards to wind resource analysis, the
larger the scale parameter is in value, the higher the mode of the distribution, which basically
means that the probability of a random wind speed value is less likely to be lower than the mode
of the distribution [29], [30].
In relation to the four offshore sites, the WAsP forecasted statistical parameters were used
as a basis for defining wind resource dispersion over the site area. Each of the meteorological
points from each site, as depicted in figures’ 11; 13; 15 and 17, were analysed. WAsP gives an
empirical relationship of roughness length and roughness elements taken from the description
given by Lettau [31]. The simple description is given as follows [31]:
(ℎ. 𝑆) (4)
𝑧0 = 0.5
𝐴𝐻
Where:
𝑧0 Roughness length (m)
ℎ Element height (m)
𝑆 Cross-sectional surface facing the wind (m2)
𝐴𝐻 Average element horizontal area (m2)
The above is more suited for terrestrial based applications as it considers various vegetation
and foliage effects in relation to the wind flow. Charnock [32] provides a description of surface
roughness phenomena over water bodies and explains that the roughness length of different
surfaces directly affects the wind speed which WAsP further defines as roughness classes. A
full list of surface roughness guidelines is illustrated in Appendix A. Charnock [32] developed
an equation which ignored the viscous effects and the surface tension of the water are ignored,
as follows:
𝑈∗ 2 (5)
𝑧0 = 𝑏
𝑔
Where:
𝑧0 Roughness length (m)
𝑏 Constant (-)

𝑈∗ Frictional Velocity (m/s)

𝑔 Gravitational acceleration (m/s2)
The WAsP tool has defined a constant roughness length of 0.0002 m for wind resource over
water bodies. During the development of the software both a constant value and equation 5 was
used. The resultant constant value proved to give good results in comparison to equation 5 and
hence maintained the constant roughness value over water bodies. The DTU Wind Energy
project [33] used onshore high-resolution meteorological measurements taken over extensive
periods to establish and develop WASA. The interpolation of results to offshore water bodies
was conducted within the WAsP software based on the onshore readings and respective surface
roughness in relation to water bodies. For the identified offshore sites, these results were used
to determine the probability distribution of the resource based on the statistical parameters
obtained from the data.
4.2.1. Site 1
Figure 19 shows the general wind characteristics for Site 1. The Weibull distribution for the
two dominant sectors of the site’s general wind direction was constructed using the wind
statistic parameters obtained via the online wind atlas synthetic data. Sectors 2 and 8 proved to
have the highest frequency and were analysed. It can be seen that Sector 2’s distribution is more
Gaussian in shape due to the k-shape factor being closer to 3. The skewness (close to 0) of
both data sets indicates a more normally distributed dataset. The mean wind speed for both the
sectors is 10.5 m/s at a 50 m hub height. It may be noted that Sector 2’s probability density
distribution depicts more probable wind speeds between 6 m/s and 15 m/s in comparison to
Sector 8’s distribution which depicts a higher probability of higher wind speeds over 15 m/s.
Figure 19 General wind characteristics – Site 1

4.2.2. Site 2
The general wind characteristics for Site 2 are shown in Figure 20. It was found that Sectors 3
and 10 were the dominant wind directions for the site. The Weibull distributions of these sectors
are shown. Sector 3 has a higher density for wind speeds of 5 m/s to 14 m/s and lower
probabilities of higher wind speeds. The distribution of wind speed in Sector 10 shows a lower
peak but higher density of probabilities between 6 m/s to 16 m/s, including higher probabilities
of wind speeds greater than 15 m/s. The mean wind speed for Sectors 3 and 10 are 9.7 m/s and
11.1 m/s respectively at a 50 m hub height.
4.2.3. Site 3
Site 3 is unique as it is located at the southern tip of the country. This region is known to have
mixing of the warm Agulhas current (east coast) and the cold Benguela current (west coast).
Along with the trade winds stemming from the South Pole towards the equator and the Coriolis
effect from the planet’s rotation, Site 3 would be expected to have cross-cutting winds. The
wind rose shown in Figure 21 shows that the most frequent directions are from Sectors 9 and
10, although the span of wind directions in comparison to the other three sites is much larger
with relatively equal frequencies which is assumed to possibly be due to the cross-winds
occurring at the site’s location. Sectors 6 and 9 were chosen for analysis. The mean wind speed
for both sectors was relatively similar with Sector 6 having a mean value of 8.22 m/s and Sector
9 having a mean value of 9.2 m/s. Comparing the two probability distributions it can be seen
that Sector 9 has higher probabilities of higher wind speeds especially between wind speeds of
12 m/s to 20 m/s, in relation to Sector 6.

4.2.4. Site 4
Figure 22 illustrates the general wind characteristics for Site 4 located west of the Northern
Cape. The dominant wind directions are Sectors 6 and 7. Comparing the two probability density
distributions, it is evident that Sector 6 experiences higher valued wind speeds compared to
Sector 7. Both sectors have shape factors greater than 3 leading to the Gaussian shape of both
distributions. The mean wind speed for Sector 6 is 9.8 m/s and Sector 7 is 7.8 m/s at 50 m hub
height. In comparison to the previous sites, Sites 4 and 3 have lower mean wind speed values
than Sites 2 and 1.

Each of the above sites were evaluated and Table 1 below encapsulates each of the sites’
wind mean speed and power density characteristics. It can be seen by evaluation of the data
table that Sites 2 and 3 have good mean wind speeds and encapsulate the most power density
of the four sites. There is a need to analyse each of the meteorological points within the area
chosen as the mean wind speed and power density varies over large areas.
Table 1 Site wind and power characteristics
Height Parameter Site 1 Site 2 Site 3 Site 4

Mean speed [m/s] 9.19 9.48 9.12 8.42
50 m
Power density [W/m²] 795.09 939.11 808.11 550.71
Mean speed [m/s] 9.25 9.55 9.11 8.50
100 m
Power density [W/m²] 852.00 1017.03 875.56 591.00
Mean speed [m/s] 9.34 9.68 9.27 8.65
200 m
Power density [W/m²] 982.27 1201.31 1032.78 684.88
4.3. Site Geographical Analysis

It is visualised that the energy which is produced offshore from any of the potential sites will
be fed to the coastal regions. In order to be as cost effective as possible, utilisation of existing
electrical integration points is advised initially (unless specific assets are required) for the most
optimum power transfer to shore. To gauge this, the distance to the site sub-integration point
(assumed to be the site central location) from the main feeder was investigated with regards to
surface elevation in respect to the standard sea surface level.
4.3.1. Site 1
Site 1 is based off the coast of Durban and has an area of 21 000 km2 as previously stated. There
are two main power incomers as shown in Figure 23, namely, Klaarwater (a) and Illovo (b) with
each having 275 kV lines respectively [18]. Figure 23 shows that the distance to the different
tie in points is roughly 200 km with the elevation level to the centroid of the site area also
included. It is noted that the deepest point is roughly 3 000 m from the ocean surface.
Figure 23 Location of possible points of connection – Site 1

4.3.2. Site 2
Figure 24 shows the distance from the various possible onshore connection points to the centre
of Site 2 which is assumed to be the central integration point of harvested offshore wind energy
for the site. The three potential connection points are Leeches Bay (132 kV), Poseidon (220
kV) and Proteus (400 kV), all of which are in the Eastern Cape. On average, the distance to
onshore connections is 409 km with the shortest distance being in relation to the Poseidon
substation. There is a distinct steep shelf drop off the Eastern Cape coastline.
4.3.3. Site 3
Figure 25 shows the two closest transmission substations, either of which could function as an
integration point for Site 3, namely, Acacia (400 kV) and Palmeit (400 kV). On average the
distance to an onshore integration point is 280 km with Palmeit being the closest at a distance
of 275 km offshore. The diagram also indicates the coastal shelf declination which extends
further out than the previous site declinations.

4.3.4. Site 4
Site 4, as seen in Figure 26, is located 527 km on average from the western coastline of South
Africa. The province/state which is closest to the site is the Northern Cape. The province is the
largest in the country but has the lowest population in the country. Limited connection points
are located along the western coast. Only two transmission connection points could be located
but there may be smaller coastal distribution feeders where connection may be possible. The
two transmission substations are Gromis (400 kV) and Juno (400 kV), each having a distance
to site of over 450 km.


The gradual evolution of offshore wind farms are a result of most regions already utilising
most of the space onshore zoned for energy production structures as well as the higher resource
potential located offshore [6]. As shown above, the depths at each of the sites are far deeper
than what traditional offshore wind turbines are subjected to. For this reason, most of the wind
farms located at such distances offshore would require a floating structure as depths in these
regions are beyond conventional anchoring.
5. CONCLUSION
This article aimed to investigate the offshore wind potential for the South African coast. Various
factors were taken into consideration before demarcating potential sites for investigation, such
as shipping routes, potential exploration oil and gas fields and possible transmission connection
points. Utilising this methodology, four potential sites were identified. The predominant wind
direction, mean wind speed, wind shear and spatial geographic information was analysed for
each site.
The wind direction for each of the sites were dependent on many factors which influence
the wind resource such as cross-wind interference and ocean currents. The wind direction did
not have any similarities between sites, with each site having its own predominant wind
directions. Site 1 had a dominant wind direction from a north easterly direction and the polar
opposite as well. Site 2’s prevailing wind direction was from the west and occasionally from a

north easterly direction. Site 3 had a large span of wind direction as it is located in a high mixing
zone of various strong wind forces. Site 4’s principle wind direction was from the south east
along the western coastline of the Northern Cape province.
Within the 50 m hub height, Site 2 showed the best potential based on the power density.
In comparison, Site 1 and Site 3 showed similar power densities to each other and Site 4
depicted the lowest power density. The general wind speed characteristics were analysed for
each of the sites and the average wind speed (taking into consideration all of the sites identified)
was 9.5 m/s with large power densities between 500 W/m2 to over 1 000 W/m2.
The distance to shore ranged from 200 km to 500 km with a steep continental shelf drop to
approximately 3 000 m deep. This means that the traditional means of offshore wind platforms
would not be adequate for anchoring the plant, and a floating offshore structure would be more
suitable. The use of a centrally located electrical energy collector hub within each site, with
possibly N – 2 contingency measures, could be useful to connect to fewer feeder bays within
the closest transmission substation onshore. This could also allow for the filtering out the power
variations from widely spaced turbines.
The study conducted shows that there is offshore wind potential off the coast of South
Africa. Energy generated by this method could assist South Africa to increase access to energy,
reduce expensive transmission line losses to coastal provinces, and assist the country transition
towards a more sustainable future energy mix in line with developed nations.
ACKNOWLEDGMENTS
We would like to acknowledge the assistance of government institutions for the assistance in
information as well as our colleagues at the university for providing direction to research.
REFERENCES
[1] Whiteman, A., Esparrago, J., Rinke, T., Elsayed, S., Arkhipova, I., Strinati, C. et al.
Renewable Energy Statistics 2018. (IRENA), Abu Dhabi, 2018.
[2] Green R. and Vasilakos, N. The Economics of Offshore Wind. Energy Policy, 39(2),
2011, pp. 496-502.
[3] Esteban Dolares, M., Javier Diez, J., Lopez Jose, S. and Negro, V. Why Offshore Wind
Energy. Renewable Energy, 36, 2011, pp. 444-450.
[4] Kaldellis J. and Kapsali, M. Shifting Towards Offshore Wind Energy – Recent Activity
and Future Development. Energy Policy, 53, 2013, pp. 136-148.
[5] Werapun, W., Tirawanichakul, Y., Kongnakorn, W. and Waewsak, J. An Assessment of
Offshore Wind Energy Potential on Phangan Island by in Southern Thailand. Energy
Procedia, 52, pp. 2014, pp. 287-295.
[6] Breton S.-P. and Moe, G. Status, Plans and Technologies for Offshore Wind Turbines in
Europe and North America. Renewable Energy, 34, pp. 2009, 646-654.
[7] Bilgili, M., Yasar, A. and Simek, E. Offshore Wind Power Development in Europe and its
Comparison with Onshore Counterpart. Renewable and Sustainable Energy Review, 15,
2011, pp. 905-915.
[8] London Array Limited. London Array. n.d. Available: http://www.londonarray.com/the-
project-3/phase-2/
[9] Da, Z., Xiliang, Z., Jiankun, H. and Qimin, C. Offshore Wind Energy Development in
China: Current Status and Future Perspective. Renewable and Sustainable Energy
Reviews, 15, 2011, pp. 4673-4684.
[10] Sun, X., Huang, D. and Wu, G. The Current State of Offshore Wind Energy Technology
Development. Energy, 41, 2012, pp. 298-312.

[11] Murthy K. S. R. and Rahi, O. P. A Comprehensive Review of Wind Resource

Assessment. Renewable and Sustainable Energy Reviews, 72, 2017, pp. 1320-1342.
[12] Sharma K. and Ahmed, M. R. Wind Energy Resource Assessment for the Fiji islands:
Kadavu Island and Suva Peninsula. Renewable Energy, 89, 2016, pp. 168-180.
[13] Lima, D. K. S., Leão, R. P. S., dos Santos, A. C. S., de Melo, F. D. C., Couto, V. M., de
Noronha, A. W. T. et al., Estimating the Offshore Wind Resources of the State of Ceará in
Brazil. Renewable Energy, 83, pp. 2015, 203-221.
[14] Kim, T., Park, J.-I. and Maeng, J. Offshore Wind Farm Site Selection Study Around Jeju
Island, South Korea. Renewable Energy, 94, pp. 2016, 619-628.
[15] Mahdy M. and Bahaj, A. S. Multi Criteria Decision Analysis for Offshore Wind Energy
Potential in Egypt. Renewable Energy, 118, 2018, pp. 278-289.
[16] Department of Wind Energy. Global Wind Atlas 1.0 [Online]. Available:
http://science.globalwindatlas.info/index.html
[17] Marine Traffic. Density Plots 2017, March. Available: https://www.marinetraffic.com/
[18] Harvard College. (2017, March). Africa Electricity Transmission Network (AICD).
Available:
https://worldmap.harvard.edu/data/geonode:africa_electricity_transmission_network_o7k.
2017, March
[19] Eskom Holdings, Eskom – Transmission Development Plan 2016-2025, 2016.
[20] Petroleum Agency South Africa. Petroleum Agency SA. 2013, 23 July. Available:
https://www.petroleumagencysa.com/
[21] International Renewable Energy Agency (IRENA), Average Wind Speed 1 km at 50m
Height DTU 2015. In Global Atlas for Renewable Energy, Wind Map, IRENA, 2015.
[22] Cunden, K. Design of a Novel Hydrokinetic Turbine for Ocean Current Power
Generation. Master of Science in Mechanical Engineering. Master’s Thesis, Mechanical
Engineering, University of Kwa-Zulu of Natal, Durban, South Africa, 2015.
[23] Pacheco, A., Gorbena, E., Sequeira, C. and Jerez, S. An Evaluation of Offshore Wind
Power Production by Floatable Systems: A Case Study from SW Portugal. Energy, 131,
2017, pp. 239-250.
[24] Sanz Rodrigo, J., Cantero, E., Garcia, B., Borbon, F., Urigoyen, U. and Lozano, S.
Atmospheric Stability Assessment for the Characterization of Offshore Wind Conditions.
Journal of Physics: Conference Series, 625, 2015.
[25] Hsu, S., Meindl, E. and Gilhousen, D. Determining the Power-Law Wind Profile
Exponent under near Neutral Stability Conditions at Sea. Journal of Applied Meteorology,
33, 1994, pp. 757-765.
[26] Azad, A. K., Rasul, M. G., Islam, R. and Shishir. I. R. Analysis of Wind Energy Prospect
for Power Generation by Three Weibull Distribution Methods. Energy Procedia, 75,
2015, pp. 722-727.
[27] Ouarda, T. B. M. J., Charron, C., Shin, J. Y., Marpu, P. R., Al-Mandoos, A. H., Al-
Tamimi, M. H. et al., Probability Distributions of Wind Speed in the UAE. Energy
Conversion and Management, 93, 2015, pp. 414-434.
[28] Wang, J., Qin, S., Jin, S. and Wu, J. Estimation Methods Review and Analysis of Offshore
Extreme Wind Speeds and Wind Energy Resources. Renewable and Sustainable Energy
Reviews, 42, 2015, pp. 26-42.
[29] Patel, M. R. Wind and Solar Power Systems, 2nd ed. New York: CRC Press, 2005.
[30] Hodge, B. K. Alternative Energy Systems and Applications. New York: John Wiley &
Sons, Inc, 2010.

[31] Lettau, H. Note on Aerodynamic Roughness-Parameter Estimation on the Basis of

Roughness-Element Description. Journal of Applied Meteorology, 8, 1969, pp. 828-832,.
[32] Charnock, H. Wind Stress on Water Surfaces. Quarterly Journal of the Royal
Meteorological Society, 81, 1955, pp. 639-640.
[33] DTU Wind Energy. WAsP 11.2. Technical University of Denmark, 1987.
APPENDIX
Appendix A
Physical z0 Roughness z0 specified in WAsP

Terrain surface characteristics
[m] Class [m]
1.5 4 (1.5 m) 1.5
>1 tall forest >1
1.00 city 1.00
0.80 forest 0.80
0.50 suburbs 0.50
0.40 3 (0.40 m) 0.40
0.30 shelter belts 0.30
0.20 many trees and/or bushes 0.20
0.10 farmland with closed appearance 2 (0.10 m) 0.10
0.05 farmland with open appearance 0.05
farmland with very few
0.03 1 (0.03 m) 0.03
buildings/trees
0.02 airport areas with buildings and trees 0.02
0.01 airport runway areas 0.01
0.008 mown grass 0.008
0.005 bare soil (smooth) 0.005
0.001 snow surfaces (smooth) 0.003
0.0003 sand surfaces (smooth) 0.003
0.0002 (used for water surfaces in the Atlas) 0 (0.0002 m) 0.0
0.0001 water areas (lakes, fjords, open sea) 0.0
“It should be noted, that in general the roughness length as applied in WAsP has to be
considered as a climatological parameter because the roughness of an area changes with
foliation, vegetation, snow cover and so on. The energy production of a wind turbine must be
determined on the basis of climatology, primarily because of the variations of the weather;
however, the seasonal variations in the local terrain characteristics can also have a profound
influence.
It is recommended that land surfaces are not assigned roughness lengths smaller than 0.003
m (3 mm); this is especially important when met. masts, reference masts or turbines are located
in such terrain with anemometer heights / heights / hub-heights more than 80 m.”

NOTES
(1) Specific production is a factor which normalises an electrical generation plant’s
production by comparing the actual generated energy with that of the installed
nameplate capacity of the generator(s). The units of measurement are kWh/kWP.
(2) A retroflection zone is an area in which the flow in question encounters specific
obstacles which cause the flow to diverge from the original path and turn on itself, either
clockwise or anticlockwise in direction.

Offshore Wind Resource Assessment Off The South African Coastline

Uploaded by

Copyright:

Available Formats

Offshore Wind Resource Assessment Off The South African Coastline

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Offshore Wind Resource Assessment Off The South African Coastline

Uploaded by

Copyright:

Available Formats

International Journal of Mechanical Engineering and Technology (IJMET)

OFFSHORE WIND RESOURCE ASSESSMENT

Figure 1 Total global renewable energy installed capacity

Figure 2 Market share of renewable capacity 2000-2016

Figure 3 Renewable energy technology growth

Figure 4 (a) Installed capacity; (b) Normalised production

Figure 5 Meteorological data points

Figure 6 Shipping routes [17]

http://www.iaeme.com/IJMET/index.asp 100 [email protected]

Figure 7 Transmission electrical grid South Africa [18]

Figure 8 Offshore oil and gas – identified sites [20]

http://www.iaeme.com/IJMET/index.asp 101 [email protected]

Figure 9 Identified potential sites

4. SITE INVESTIGATION & ANALYSIS

http://www.iaeme.com/IJMET/index.asp 102 [email protected]

4.1. Wind Direction

Figure 10 Potential – Site 1

Figure 11 Site 1 – Wind rose plots

http://www.iaeme.com/IJMET/index.asp 103 [email protected]

Figure 12 Potential – Site 2

Figure 13 Site 2 – Wind rose plots

http://www.iaeme.com/IJMET/index.asp 104 [email protected]

Figure 14 Potential – Site 3

Figure 15 Site 3 – wind rose plots

http://www.iaeme.com/IJMET/index.asp 105 [email protected]

Figure 16 Potential – Site 4

Figure 17 Site 4 - wind rose plots

4.2. Wind Speed

http://www.iaeme.com/IJMET/index.asp 106 [email protected]

𝑢∗ Frictional velocity (m/s)

Figure 18 Extrapolated wind shear profiles

http://www.iaeme.com/IJMET/index.asp 107 [email protected]

http://www.iaeme.com/IJMET/index.asp 108 [email protected]

𝑈∗ Frictional Velocity (m/s)

Figure 19 General wind characteristics – Site 1

http://www.iaeme.com/IJMET/index.asp 109 [email protected]

Figure 20 General wind characteristics – Site 2

http://www.iaeme.com/IJMET/index.asp 110 [email protected]

Figure 21 General wind characteristics – Site 3

Figure 22 General wind characteristics – Site 4

http://www.iaeme.com/IJMET/index.asp 111 [email protected]

Table 1 Site wind and power characteristics

Height Parameter Site 1 Site 2 Site 3 Site 4

4.3. Site Geographical Analysis

Figure 23 Location of possible points of connection – Site 1

http://www.iaeme.com/IJMET/index.asp 112 [email protected]

Figure 24 Location of possible points of connection – Site 2

http://www.iaeme.com/IJMET/index.asp 113 [email protected]

Figure 25 Location of possible points of connection – Site 3

http://www.iaeme.com/IJMET/index.asp 114 [email protected]

Figure 26 Location of possible points of connection – Site 4

http://www.iaeme.com/IJMET/index.asp 115 [email protected]

http://www.iaeme.com/IJMET/index.asp 116 [email protected]

[11] Murthy K. S. R. and Rahi, O. P. A Comprehensive Review of Wind Resource

http://www.iaeme.com/IJMET/index.asp 117 [email protected]

[31] Lettau, H. Note on Aerodynamic Roughness-Parameter Estimation on the Basis of

Physical z0 Roughness z0 specified in WAsP

http://www.iaeme.com/IJMET/index.asp 118 [email protected]

http://www.iaeme.com/IJMET/index.asp 119 [email protected]