Proceedings of the ASME 2013 32nd International Conference on Ocean, Offshore and Arctic Engineering

OMAE2013
June 9-14, 2013, Nantes, France

OMAE2013-10660

SCOUR ASSESSMENT AND MEASUREMENTS FOR PILE-SUPPORTED WIND

TURBINE FOUNDATIONS
Bruno Stuyts David Cathie
Cathie Associates Cathie Associates
Diegem, Belgium Diegem, Belgium

Yi Xie
Advanced Geomechanics
(formerly Cathie Associates)
Perth, WA, Australia

ABSTRACT
With the rapid development of offshore wind energy in
Europe, a large number of piled structures are being installed. INTRODUCTION
In areas with sandy seabed conditions, erosion of sediment by Offshore wind energy is rapidly developing into a major
the actions of wave and current can negatively influence source of renewable energy in Western Europe with up to 3GW
foundation capacity. An accurate prediction model of scour of installed capacity in April 2011 and a further 2.5GW under
around the piles is therefore required. Well-accepted scour construction [3]. Several North Sea wind farm sites are
prediction methods exist; both for the equilibrium scour depth positioned in glacial deposits with loose to medium dense
and the time scale of scour [1] around single piles. These layers of cohesionless soil at the surface. Piled foundations
standard formulas have been combined with metocean data and installed in such soil conditions will inevitably be subject to
a hindcasting model to calculate the expected scour depth scour [4].
around piles of wind turbine tripod foundations. Other causes Scour assessment methods were developed in the past for
of scour, such as pile-pile interaction, effect of proximity of offshore Oil & Gas infrastructure and can be applied to
structural members to the seabed, and seabed mobility were foundations for offshore wind turbines. This paper shows how
also assessed in order to determine the amount of global scour calculated equilibrium scour depths and time scales of scour
to be considered. The scour predictions were compared to can successfully be combined with a hindcasting model to
measurements taken at an offshore wind turbine foundation at obtain good predictions. An example for a piled tripod in the
Park Alpha Ventus (PAV) in the German North Sea [2]. The German Sector of the North Sea is presented.
data showed very good agreement with the measured scour For piled foundations, scour will have an impact on the
around the piles. Both the equilibrium scour depth and time axial pile foundation capacity. The method proposed by the
scale of scour were well predicted using the hindcasting model. American Petroleum Institute (API) [5] is compared to
The measured scour below the central column of the tripod numerical modeling results. The numerical model takes the
structure exceeded expectations; this is believed to be due to a scour pits around each of the piles into account and determines
pumping effect during storm episodes. Finally, the effect of the effect of scour on the effective stresses at the pile-soil
scour on the vertical effective stress around the tripod piles was interface. The results show that the API recommendation is a
assessed with a finite element model. Local scour had an conservative approximation.
important effect while scour below the centre of the structure
had a much more limited effect. Considering the combined
effects of multiple pile interaction, scour below the central
column, and making an allowance for seabed mobility, an
equivalent global scour depth for pile capacity calculations was
established.

1 Copyright © 2013 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 NOMENCLATURE θcr = 0.05 to 0.06. Alternatively, a formula dependent on grain
d50 Median grain size size can be used [6]. For the example in this paper, this
s Specific gravity of solid sand particles corresponds to a critical bed shear velocity u*cr of 0.013m/s.
 Shields parameter The relationship between bed shear velocity u* and the
u* Bed shear velocity depth averaged current velocity Ū is given in Equation 2 [6]
g Acceleration of gravity
⁄
θcr Critical Shields parameter ‫ ∗ݑ‬1 ݀50 1 7
= ∙൬ ൰
u*cr Critical bed shear velocity ܷഥ 7 ℎ (2)
Ū Depth-averaged
averaged current velocity Where h is the water depth. The water depth at the site in
Ūcr Critical depth-averaged
averaged current velocity this example
le is 30m. A critical depth averaged current velocity
h Water depth Ūcr = 0.5m/s was thus calculated.
calculated This current speed is
Hs Significant wave heights compared to the current data at the site in one of the subsequent
KC Keulegan-Carpenter number sections.
umax Maximum value of orbital velocity
B Bracing diameter
Mobility due to waves
D Pile diameter
Waves in sufficiently shallow water (h < 10.Hs) produce an
Tz Wave period
oscillatory velocity at seabed [6].
Seq Equilibrium scour depth
induced scour is primarily governed by the
The wave-induced
t time
Keulegan-Carpenter
Carpenter number, KC in Equation 3 [1]
Tchar Characteristic time scale of scour
u maxTz
KC  (3)
SEDIMENT MOBILITY D
Where umax is the maximum value of the orbital velocity,
velo D
Seabed material is the pile diameter and Tz is the wave period. A wave period
Cohesionless soils are subject to erosion due to the action dependent on Hs was used for the calculations.
of currents and waves. The medium grain size (d50) of the A graph of KC vs. significant wave height, Hs is shown in
particles and the specific gravity of the solid particles (s) are the Figure 1 for
or 30.0m water depth. DNV [1] state that for KC < 6,
principal input parameters for sediment mobility calculations. no scour hole is formed. The chart shows that Hs must exceed
In the example for a foundation at PAV presented in this 8m before wave-induced
induced scour takes place.
paper, the seabed consisted of fine to medium sand with a
median grain size d50 = 0.13 – 0.29mm. A mean value of Figure 1 – KC vs. Hs for 30m water depth
0.20mm was used for all calculations. A specific gravity s=2 s=2.6 14
was derived from geotechnical classification
tion tests.
Keulegan-Carpenter number, KC

The bedform of the sand is also important and relevant for 12

the scour assessment. The geophysical survey performed at the
Threshold for wave
site showed a featureless seabed, with little or no local re relief. 10
induced scour
This also suggested that the seabed was not highly dynami dynamic.
Hence, significant changes in seabed elevation due to sand 8
migration were not anticipated. 6

Mobility due to currents 4

When currents are acting on a flat seabed, the sandy seabed
material will be immobile until a certain threshold current 2
ed. The Shields parameters θ, is defin
velocity is reached. defined in
Equation 1. 0
0 2 4 6 8 10 12
Significant wave height, Hs [m]
(‫) ∗ݑ‬2
ߠ=
݃(‫ݏ‬− 1)݀50
(1)
Where u* is the bed shear velocity and g is the acceleration METOCEAN DATA
of gravity (9.81m/s²).
The critical value of the Shields parameter, θcr, is the value Current
of the Shields parameter for which sediment motion is initiated. Insight
nsight into the current regime is provided directly by field
DNV [1] state that the critical value of the Shields parameter is data measurements. Currents have been measured since 2003 at

2 Copyright © 2013 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 the FINO1 platform, which is located in the vicinity of the site Figure 3 – Sinusoidal fit to current data at 8m water
considered in this paper. The current speeds and directions depth for FINO1
collected during a 2.5 year period are described by Herklotz
[7].. The current measurements at 8m water depth are shown in
Figure 2.. The water depth at the example location is nearly
identical to the water depth at FINO1. The orange line at the
bottom of the figure represents the current speed, the black line
the current direction.

Figure 2 – 2005 current statistics at 8m water depth

for FINO1 [7] Waves
The wave climate was also measured at FINO1. The waves
measured during the storm "Tilo" are described in [8] (Figure
4). The chart shows that during the extreme storm, the
significant wave height exceeded 8m during a period of
approximately 1 hour. Figure 1 suggests therefore that wave-
wave
induced scour will therefore be limited during the lifetime of
the wind turbine foundations.

Figure 4 – Significant wave heights measured at

FINO1 during storm "Tilo", November 2007 [8]

SCOUR CALCULATIONS

Equilibrium scour depth epth for single piles

The equilibrium local scour depth,
de Seq, is given by DNV
[1] for a single cylindrical pile as the expression in Equation 4.
The relative importance off the current and waves is governed
by the Keulegan-Carpenter
Carpenter number KC.

 1.31  exp 0.03KC  6 

S eq
(4)
In order to provide input for the model for scour D
development with time, a sine-wave
wave was fitted to the peaks of
the current velocity at 8m depth as shown in Figure 3. The This formula predicts no scour when KC < 0. However,
sinusoidal fit corresponding to the spring tide/neap tide cycle is DNV state that for steady current KC→∞.KC In this case the
shown in yellow. An average current velocity of 0.4m/s in equilibrium scour depth is 1.3D. Several other formulations
combination with an amplitude of 0.2m/s fits the data well. exist but these all refer back to the equilibrium scour depth of
1.3D for steady current conditions. Note that for large diameter
piles, this formula must be used with caution [1].
The scour depth in steady current conditions is given by
Whitehouse [9] in Equation 5 based on the Shields parameter θ.
Equation 5 is based on scale model tests [9].

3 Copyright © 2013 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 S eq A and B are calibration parameters which were determined
 1.3 when    cr to be A = 0.014 and B = -1.29 [10].
D
(5)
S eq     Hindcasting method for single pile scour
 1.3 2  1.0  when 0.25  1
D   cr   cr Rudolph et al [11] propose a scheme for hindcasting of
local scour development based on available metocean data. A
given period of time is split up into a number of time
When θ/θcr<0.25, there is no scour. Figure 5 shows the
increments, Δt (Δt equals the resolution of the metocean data),
variation of scour depth with flow speed for the scale model
and for each time increment, the current speed, water depth and
tests.
wave orbital velocity is computed. Based on the properties of
the bed material and the hydrodynamic parameters, the
Figure 5 – Variation of scour depth with flow speed
equilibrium scour depth and characteristic scour depth can be
[9]
determined for each time step.
For current dominated conditions, the equilibrium scour
depth is computed according to Equation 5 and the
characteristic time scale is given by Equation 7. For each time
step, the computed equilibrium scour depth is compared to the
scour depth at the end of the previous time increment (Sn). If
the equilibrium scour depth (Seq,n+1) is larger than the scour
depth for the previous time increment, scour is allowed to
develop according to Equation 8. If the equilibrium scour depth
is smaller than the scour that has already developed, the scour
depth remains constant. The process of back-filling of the scour
hole is therefore ignored. This is a conservative assumption
which can be justified at locations where large-scale sediment
transport is not observed.

S n 1  S n  S t 
  t  (8)
Time scale of scour development for single piles S t   S eq ,n 1 1  exp   if S eq ,n 1  S n
  T 
Scour will only start to develop when the currents and/or  char ,n 1 
waves are intense enough to cause seabed mobility. In current
dominated conditions, scour can take months or years to Detailed numerical data of the hydrodynamic parameters
develop. In wave dominated conditions, equilibrium is reached (current velocity, significant wave height and water depth) was
relatively quickly (order of magnitude of a few hours). not available at the windfarm site under consideration.
Scour development is not a steady process. The current Therefore, an indicative simulation of the scour development
data in Figure 3 shows that the threshold current velocity is was performed using the sinusoidal fit to the current data from
only exceeded during peaks of tidal current velocity. Also, Figure 3. Equation 8 is applied to the sinusoidal time series and
wave-induced scour will only develop during extreme storms as the calculated evolution of scour depth is presented in Figure 6.
shown in Figure 1. In this example, the pile diameter D=2.5m.
The characteristic time scale of local scour around single The sensitivity to the grain size was also checked using a
piles as well as a formula for scour development in current- low estimate d50 = 0.13mm, a best estimate d50 = 0.20mm and a
dominated conditions is given in [1] and [10]. Equation 6 high estimate d50 = 0.29mm. In all cases, an equilibrium scour
shows the formula for scour development while the depth of 3.25m is reached. For the best estimate grain size, the
characteristic time scale, Tchar, is given in Equation 7. scour hole reaches and equilibrium depth after approximately
200 days. For the low estimate grain size, the scour
  t  development takes approximately 20% more time. The scour
S t   S eq 1  exp  
 (6) development for the high estimate grain size is ~15% quicker.
  Tchar  Note that only current-induced scour was considered in this
calculation due to the relatively deep water at the wind farm
site.
T *  A B
T *D2 (7)
Tchar 
g s  1d 50
3

4 Copyright © 2013 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 Figure 6 – Prediction of scour depth development However, uncertainty about the possibility of seabed
based on sinusoidal fit to current velocity lowering can be considered implicitly in the selection of a
3.5 conservative global scour
cour value for design.
design
3.0
Scour depth [m]

2.5
2.0
SCOUR PREDICTIONS VS.. FIELD MEASUREMENTS
Low estimate grain size
1.5
Best estimate grain size
Tripod piles
1.0
Scour was measured around the tripod piles at location
0.5 High estimate grain size AV7 of Park Alpha Ventus (PAV) with echosounders and
0.0 through a bathymetric survey [2].
[2] A drawing of the tripod
0 100 200 300 400 foundations used at Alpha Ventus is shown in Figure 7.
Time [days]

Figure 7 – Tripod foundation used at Park Alpha

Ventus [2]
Global scour
Global scour (also called
led dishpan scour) is caused by the
change in flow velocity and turbulence due to the close
proximity of several piles and/or other structural elements. For
engineering design, global scour is assumed to consist of an
overall lowering of the seabed around the entire structure which
in turn reduces the effective vertical stress around the piles.
Global scour also increases the depth where no lateral
resistance is considered. However, axial pile capacity is
particularly affected by loss of vertical stress aro
around the pile
and therefore global scour is assessed accounting for the Fixed echosounders were placed along the pile sleeves and
average stress reduction that is appropriate for a given underneath the central column. The scour measurements from
structure. Scour induced by the presence of structural elements this test field provided a means to validate the methodology
near the seabed does not necessarily result in general seabed outlined in this paper. The measured and an the predicted scour
lowering
wering but can contribute to the reduction in vertical effective depth around the piles are plotted in Figure 8. The wave climate
stress. Therefore, global scour is a means by which the average measured at the site is also plotted for information. Red dashed
effect of local scour can be accounted for, as well as a method lines represent extreme storm events.
for dealing with general lowering of the seabed. The figure shows that the predictions agree very well with
For this study, global scour has been considered to arise the measured data. As predicted, d, the scour around the tripod
from the following sources: piles reaches an equilibrium depth of approximately 3.0-3.5m
3.0
between 150-200200 days after installation. It should also be noted
• Effect of turbulence and scour induced by multiple that scour depth is not uniform around the tripod pile. Scour is
piles; most severe at the sensor ES 6 which
whi is most exposed to the
• Effect of tripod bracing and central column; dominating current direction.
• Effect of seabed sediment mobility. The data shows no clear correlation between storm periods
and scour depth. No notable increase in the scour depth is
Scour due to the interaction of multiple piles is well measured during periods with high significant wave height. The
documented [10] and can be quantified based ed on experimental severest storms occurredrred after reaching equilibrium scour
data for pile groups although test data for the scour depth, notable increase were therefore not expected.
development around three piles in a triangular configuration in It should be noted that at the end of August 2010 and the
current-dominated
dominated conditions is not published to dat
date. end of October 2010, there is a sudden change in the measured
The effect of foundation bottom struts (bracing) can be scour depth. It is not clear whether this sudden su change
treated as a horizontal tubular member (using methods corresponds to sudden change in the scour depth or whether it
developed for spanning pipelines) with a certain offse
offset from the is due to equipment malfunctioning.
seabed. Nevertheless, these phenomena are not documented
extensively in the literature and therefore, predictions will be Central column of tripod foundations
much less accurate than the predictions for single pile scour. Scour below the central column was also measured with
Seabed sediment mobility appears to be limited (no sand echosounders [2].. The measured scour depth is plotted in
waves or obvious bedforms). Therefore, no quantification of Figure 9. The wave climate
ate measured at the site is again plotted
general seabed lowering has been attempted. for information.

5 Copyright © 2013 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 Figure 8 – Comparison of measured and predicted Figure 9 – Measured scour depth below the central
scour depth around tripod piles (measurement data tripod column (measurement data from [2])
from [2])

6 Copyright © 2013 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 The data show that there is no steady-state
state scour depth Figure 10 – Extent of scour holes around tripod piles
reached for the central column. Most of the scour has
developed during the first 4-55 months after installation.
Afterwards, scour continues to increase slowly at a rate of
approximately
oximately 1m/yr from 5.0m at the start of 2010 to 6.0m at
the end of 2010. The scour below the central column does not
yet seem to have an influence on thee scour around the tripod
piles as shown in [2].
Comparison with the wave data shows thatt there seems to
be a correlation between periods of extreme weather and scour
development below the central column. This might be due to
the central column being open at the bottom leading to The figure shows that the local scour holes would interact
significant pumping action in periods with high waves. These to some extent with the scour hole below the central column.
pumping
mping effects were also shown in model tests documented in However, the seabed lowering due to scour below the central
[12].. However, equipment malfunctioning cannot be ruled out column does not extend to the location of the piles.
piles The effect
either. Pumping effects could be avoided on future projects by off the different local scour holes on the vertical effective stress
fitting flow diverters,, which ensure water flows out horizontally is investigated in the subsequent section.
rather than vertically, to the bottom om of the tripod central
column.
Influence of local scour holes around piles
EFFECT OF SCOUR ON PILE CAPACITY A linear elastic finite element model was set up in Abaqus
One of the main impacts of scour for pile capacity is to to quantify the reduction in effective stress due to the presence
reduce the vertical effective stress and therefore the radial stress of local scour holes. The soil was model as a linear elastic
on the piles which governs the shaft resistance. For this reason, material with a Young's modulus of E' = 100GPa, a Poisson's
the effect of scour on effective stress reduction is evaluated. tive unit weight of γ' = 10kN/m³.
ratio, v' = 0.3 and an effective
A locall scour hole with a depth of 3.2m, a radius of 8.1m
Lateral extent of scour holes and 25° side slopes was modelled to obtain the vertical
In order to model the actual scour shape in a finite element effective stress around the pile after scouring. A contour plot of
model, the lateral extent off local scour holes should be the vertical effective stress is shown in Figure 11.
determined. DNV [1] recommendss that the lateral extent can be The effect of the scour hole on the vertical effective stress
calculated according to Equation 9: at the location of the pile-soil
soil interface is shown in Figure 12.
Since there is no soil present, the vertical effective stress is zero
D S eq down to the equilibrium scour depth. At greater depths, an
r  (9) overburden recovery zone exists where the vertical effective
2 tan  stress increases back to the undisturbed value. This is in
accordance with the recommendations given in [5].
Where r is the radius of the scour hole and φ is the angle of
repose of the soil. [13] mentions that the angle on tthe Figure 11 – Vertical effective
ffective stress due to local scour
downstream side is equal to 1/2 to 2/3 of the internal friction around tripod piles
angle, φ', of the sand. On the upstream side, the angle is larger.
Assuming φ' = 37.5°, the angle of repose (2/3φ') would be φ =
25°. The analyses presented in this paper are for an angle of
repose of 2/3φ'.. The seems to be in line with the multibeam
survey results presented in [2].. More detailed analysis is
possible where the effect of waves and currents on the
submarine slope stability is studied [14] but this is beyond the
scope of this paper.
The lateral extent of the scour holes around the foundation
piles and under the central column is shown in Figure 10
assuming 5.0m 0m scour under the central column, and an
equilibrium scour depth of 3.2m for the piles.

7 Copyright © 2013 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 Figure 12 – Vertical effective stress profiles due to Figure 14 – Vertical effective stress profiles due to
local scour around tripod piles local scour around tripod piles and bracing
Vertical effective stress [kPa] Vertical effective stress [kPa]
0 -100 -200 -300 0 -100 -200 -300
0 0

5 5
Depth below mudline [m]

Depth below mudline [m]

10 10

15 15

20 20

Undisturbed seabed
Undisturbed
25 0° from bracing
25
Influence of local scour 90° from bracing
hole 180° from bracing
30 No bracing
30

The vertical effective stress for an undisturbed seabed is

also shown for information.
Influence of bracings Figure 14 shows that the influence of the bracing channel
The scour hole ole below the bracing was modeled as a on the vertical effective stress is negligible compared to the
channel with a bottom width equal to the largest bracing influence of the local scour hole. Therefore, it
i is not necessary
diameter (B = 3.82m), a depth of 1.7m and side slopes of 25°. to consider the component of global scour due to the bracings.
The depth of this channel was obtained by considering an
analogy with a free-spanning pipeline. A contour plot of the
vertical effective stress in the model is shown in Figure 13. Influence of central scour hole
The effect of the bracing scour channel on the vertical Even though a flow diverter might be fitted below the
effective stress at the pile-soil
soil interface is shown in Figure 14. central column of the tripods, the effect of possibly having
The figure shows the vertical effective stress at three locations: 5.0m of scour below the central column was analyzed to
 Pile-soil
soil interface facing the bracing channel (0°)
(0°); quantify
tify the reduction in vertical stress and therefore how much
 Pile-soil interface on the opposite side of the bracing global scour is needed to account for the central scour, should it
channel (180°); occurs.
 Pile-soil
soil interface between the two previous locations The scour below the central column was modeled as a hole
(90°). with a depth of 5.0m, a radius of 13.0m and 25° side slopes.
Figure 10 shows that the scour hole below the central column
Figure 13 – Vertical effective stress due to local overlaps to a certain extent with the local scour holes around
scour around tripod piles and bracing the tripod piles.
The contour plot of vertical cal effective stress is shown in
Figure 15.. The figure shows that the central column will cause
a reduction of the vertical effective stress which will also
influence the tripod piles to some extent.
The vertical effective stress around the tripod piles is
shown in Figure 16, and was again quantified at three locations:

8 Copyright © 2013 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

  Pile-soil
soil interface facing the central column (0°); Summary
 Pile-soil
soil interface on the opposite side of the central The influence of local scour around the piles and scour
column (180°); below the bracings and central column was quantified with a
 Pile-soil
soil interface between the two previous locations linear elastic finite element model,
model in order to understand the
(90°). change of the vertical effective stress at the pile location.
location The
different scour holes were modeled ed explicitly and the vertical
Figure 15 – Vertical effective stress
ess due to local effective stress at the pile-soil
soil interface
interf was determined at
scour around tripod piles and central column different locations around the circumference of the pile. The
results of the different contributions are shown in Figure 17
which indicates that vertical effective stress reductions of less
than 2kPa are likely to be caused by bracing and central column
scour.
The combination of 3.2m of local scour around the tripod
piles in combination with 5.0m below the central column
causes the highest reduction in vertical effective stress.
However, if the flow diverters are fitted below the central
column, the development of 5.0m of scour below the central
column is unlikely.
Therefore, considering possible contributions to scour
below the tripod from hydrodynamic interaction between piles,
the effect of scour below the tripod bracing and central column,
and making an allowance for sediment mobility and seabed
level change, a global scour depth of 0.8m is proposed. Figure
17 shows that the design profile according of vertical effective
Figure 16 – Vertical effective stress profiles due to
stress derived according to [5] (3.2m local scour plus 0.8m
local scour around tripod piles and central column
global) provides a sufficient safety margin, even for the case
with a 5.0m deep scour hole below the central column.
Vertical effective stress [kPa]
0 -100 -200 -300 Figure 17 – Vertical effective stress due to local scour
0 and global scour with design profile
p
Vertical effective stress [kPa]
5 0 -100 -200 -300
0
Depth below mudline [m]

10
5
15
10
Depth below mudline [m]

20
15
25
20
30 Undisturbed
25
0° from central column
90° from central column
180° from central column 30
Undisturbed
No central scour hole
Local scour only
Influence of central column (0°)
(0
The effective stress profile without the influence of the Influence of bracing (0°)
(0
Design profile
central column is also shown in Figure 16. The figure shows
that the central scour hole would cause a higher reduction in
vertical effective stress than the tripod bracings. A component
of global scour should therefore be considered to account for
this reduction.

9 Copyright © 2013 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 CONCLUSIONS [6] R. Soulsby, “Dynamics of Marine Sands,” Thomas
A scour depth prediction model was established in this Telford, 1997.
paper. The prediction model was based on the work by [10] [7] K. Herklotz, “Oceanographic Results of Two Years
and [11]. The model predictions were compared to the scour Operation of the First Offshore Wind Research Platform in
depth measurements at windfarm site Park Alpha Ventus (PAV) the German Bight - FINO1,” DEWI Magazin, vol. 30,
[2]. 2007.
Scour around the foundation piles is expected to reach [8] O. Outzen, K. Herklotz, H. Heinrich, and C. Lefebvre,
equilibrium after approximately 175 days and the local scour “Extreme Waves at FINO1 Research Platform Caused by
depth is not expected to exceed 1.3 times the pile diameter (i.e. Storm ‘Tilo’ on 9 November 2007,” DEWI Magazin, vol.
3.2m) based on both model predictions and measurements at 33, 2008.
PAV. [9] R. Whitehouse, Scour at marine structures: A manual for
Other causes of scour, such as pile-pile interaction, effect practical applications. Inst of Civil Engineers Pub, 1998.
of proximity of bracing and central column to the seabed, and [10] B. M. Sumer and J. Fredsøe, The mechanics of scour in the
seabed mobility were also assessed in order to determine the marine environment. World Scientific Publishing
amount of global scour to be considered. The pile-pile scour is Company Incorporated, 2002.
expected to be much less than 0.8m while the proximity of the [11] D. Rudolph, T. C. Raaijmakers, C. J. M. Stam, and W. Op
bracing has the largest effect theoretically with local scour on den Velde, “Evaluation of scour development around
the order of 1.5m. At PAV, about 5.0m scour was measured offshore monopiles based on measurements in the Q7
below the central column but this is not expected to occur if the windpark,” presented at the European Offshore Wind
fitting of a flow diverter is fitted to the bottom of the column. Conference, Berlin, 2007.
Moreover, the measured maximum scour (3.3m) around the [12] A. Stahlmann and T. Schlurmann, “Physical Modeling of
piles at PAV includes all scour effects. Scour around Tripod Foundation Structures for Offshore
The effect of scour on the vertical effective stress around Wind Energy Converters,” COASTAL ENGINEERING, p.
the tripod piles was assessed with a linear elastic finite element 2, 2010.
model. Local scour has an important effect while scour below [13] M. Achmus, Y.-S. Kuo, and K. Abdel-Rahman, “Numerical
the bracing and central column a much more limited effect. Investigation of Scour Effect on Lateral Resistance of
Should up to 5m scour below the central column occur, the Windfarm Monopiles,” in Proceedings of the Twentieth
effective stress reduction would be equivalent to approximately (2010) International Offshore and Polar Engineering
0.2m of global scour. Therefore, considering the combined Conference, Beijing, China, 2010, pp. 619–623.
effects of multiple pile interaction, possible scour below the [14] J. Bubel and J. Grabe, “Stability of submarine foundation
central column, and making an allowance for seabed mobility, a pits,” in SUT OSIG, London, 2012, pp. 347–354.
global scour depth of 0.8m is recommended.
This global scour allowance provides an additional margin
where the pile is assumed unsupported laterally.
Scour around the tripods should be monitored periodically
and if scour exceeds the design values, the impact on pile
stability and stiffness should be assessed to determine if scour
remedial measures are needed.

REFERENCES
[1] DNV-OS-J101, DNV-OS-J101 - Design of offshore wind
turbine structures. Det Norske Veritas, 2010.
[2] M. Lambers-Huesmann and M. Zeiler, “Untersuchungen
zur Kolkentwicklung und Kolkdynamik im Testfeld Alpha
Ventus,” in Veroffentlichungen des Grundbauinstitutes der
Technischen Universitat Berlin Heft Nr. 56, Berlin, 2011.
[3] PwC, “Offshore proof Turning windpower promise into
performance,” PwC offshore windpower survey – based on
field research conducted by GBI, 2011.
[4] DECC, “Dynamics of scour pits and scour protection -
Synthesis report and recommendations (Milestones 2 and
3),” The Department of Energy and Climate Change, 2008.
[5] API RP GEO, “API RP2 GEO Geotechnical and
Foundation Design Considerations,” 2011.

10 Copyright © 2013 by ASME

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use