ICSE6 Paris - August 27-31, 2012

F.H. Dixen et al.

ICSE6-202

Scour Development Around Offshore Wind Turbine Foundation:

Field Measurement & Analysis
Figen Hatipoglu DIXEN1, Martin DIXEN2 , Jan PEDERSEN3 and Jesper Fris DAHL4
1

Senior Scour Specialist/ DONG Energy

Nesa All 1, 2820 Gentofte, Denmark - e-mail:figdi@dongenergy.dk
2

Hydrodynamicist/ DHI
Agern All 5, 2970 Hrsholm, Denmark - e-mail:mdi@dhi.dk
3

Lead Specialist/ DONG Energy

Kraftvaerksvej 53, 7000 Fredericia, Denmark - e-mail:janpe@dongenergy.dk
4

Manager/ DONG Energy

Nesa All 1, 2820 Gentofte, Denmark - e-mail:jefda@dongenergy.dk

Scour development affects the stability and dynamic behaviour of an offshore wind turbine foundation. In order to
estimate the scour hole characteristics around a foundation, the reaction of the seabed, both locally and globally, to
hydrodynamic action together with the hydrodynamic climate has to be known over long time span.
This paper presents the results of an analysis of long-term scour measurements made around an existing monopile
foundation exposed to wave and current action. Measurements were realized over six months. Maximum scour depth
and scour length in longitudinal direction were measured. Local metocean parameters were used to correlate
environmental conditions and scour response. Prediction formulae for scour development were evaluated against the
measurements.

Key words
Scour, monopile, offshore, field measurement, wave, current
I

INTRODUCTION

A large number of offshore wind farms are being planned. Many of these will be placed on a seabed
consisting of movable material such as sand and silt at water depths between 10 and 30 m. This is a range
where the wave and current interaction with the foundation has a significant effect on the total load. Being
able to predict the exact scour/erosion depth will make the maximum load and fatigue estimations much
more accurate, which can lead to simpler and cheaper designs; e.g. without scour protection. This will reduce
investments and therefore make offshore wind turbines a more competitive and viable alternative. Scour has
been studied extensively during the past 10 to 15 years (Hoffmans and Verheij, 1997, Whitehouse, 1998,
Sumer and Fredse, 2002). However, work on the development of scour holes over time periods in the order
of magnitude of months/years, is sparse. No long-term site measurements have been done to investigate the
temporal variation of the scour depths.
In the present study, a combined dataset of metocean parameters and scour measurements obtained from the
turbine foundation with a diameter of 4.7 m which was installed November 2008 in Gunfleet Sands Wind
Farm provided site specific validation of how much variation in scour depth can be expected. Data obtained
1.5 years after installation of the foundation is compared with the results of mathematical model to see how
well that model captured the variation.

1441

ICSE6 Paris - August 27-31, 2012

II

F.H. Dixen et al.

MONITORING

A scour monitoring system which consists of two Imagenex 881 sonars and an ADCP was used and data
acquisition took place remotely (Fig. 1a). Sonars were used to capture the spatial and temporal variability of
scouring. The installation of two sonars was performed with one sonar on either side of the monopile, each
covering an area of more than 180 degrees. Full capture of the area obtained by combining a scan from each
of the sensors. This is illustrated above in Figure 1b, where the yellow rectangles show that the sonars have
been mounted with an approximate heading of 45 ( northeast of the monopile) and 225 degrees (
southwest of the monopile), respectively. The accuracy of the scour measurements was 0.1 m.
Oceanographic parameters were measured with ADCP.

Figure 1a : Measurement setup side view.

Figure 1b : Measurement setup - plan view.

Measurements were carried out for 195 days in the period of June - December 2010. 122 days of scour data
and 39 days of oceanographic data were collected. ADCP stopped working after two months of data
collection and missing ADCP data (wave/current data) for the period of interest was established based on a
DHI hindcast model for the Outer Thames Estuary, where the wave and current data was extracted from the
location of the wind turbine. Furthermore, a tide analysis was performed based on the measured data and the
hindcast data in order to predict the tidal water levels and current from the date of installation of the
monopile. The measured wave/current data and the ones modelled by DHI was in good agreement

within 5%, although some of the peak events were underestimated with approximately 8-10%.

1442

ICSE6 Paris - August 27-31, 2012

III

F.H. Dixen et al.

ENVIRONMENTAL CONDITIONS

Current dominated conditions are prevailed at the location. The main direction of the tidal current is
approximately 70-250N. The behaviour of waves is tidally modulated as can be seen in Figure 2, which
shows a plot of time-series of wave heights during the period from 26th October to 11th November. The
accompanying water level and tidal current data are also shown. The maximum measured depth-averaged
tidal current during the spring tide is approximately 1.1 m/s. Water depth is -11.4 m MSL and peak period of
waves is 8.3s. The seabed material consist of medium sand (d50=0.2 mm).
IV

RESULTS

Measurement results indicate a clear trend and a significant wave pattern with a period of approximately
two weeks ( Fig. 3). This pattern can be attributed to tidal variation. Local scour depth varied between 1.53D
-1.7D during the measurement period.

Figure 2: Time-series from hindcast data

Figure 3 : Temporal variation of measured scour depth.

1443

ICSE6 Paris - August 27-31, 2012

F.H. Dixen et al.

Fig. 4 shows variation of the scour hole profiles during the measurement period along the three different
directions. According to these profiles, the deepest and widest scour hole formed in the main tidal direction
(70-250N). The hole had a typical cross-sectional profile for unidirectional flows and the downstream slope
was milder than the upstream slope. Scour hole was symmetrical in the direction of 0-180N.

0 -180 N

0
1
2
3
4
5
6
7
8

Depth (m)

30

25

20

15

10

10

15

20

25

70 -250 N

0
1
2
3
4
5
6
7
8

120 -300 N
0
1
2
3
4
5
6
7
8
25

20

15

10

0
(m)

10

15

20

25

30

35

MATHEMATICAL MODEL

In order to formulate the time evolution of scouring, a mathematical model has been used. The scour
development around a pile under steady flow conditions has been defined by the following formula
[Whitehouse, 1998] :

t
S (t ) S eq 1 exp
T

1444

(1)

ICSE6 Paris - August 27-31, 2012

F.H. Dixen et al.

where Seq is the equilibrium scour depth and T is the time scale of the scour process for the given
environmental condition. The formula for equilibrium scour depth for combined wave and current has been
given as [Sumer and Fredse, 2002] :

S Sc
1 exp AKC B
D D
3 2. 6
A 0.03 U cw
4
B 6 exp 4.7U cw

KCB

(2)
(3)
(4)

where Sc is the equilibrium scour depth in current alone, KC is the Keulegan-Carpenter number, D is the pile
diameter and Ucw is given by

U cw

Uc
Uc Um

(5)

where Uc is the current velocity and Um is the maximum wave induced velocity at the bed. The normalized
time scale, T* for scour in current alone and waves alone can be calculated by
*

Tcurrent

1 h 2. 2

2000 D

KC
*
Twaves
10 6

(6)

(7)

where is the Shields parameter, and h is the water depth.

For combined current and waves, no equation exists for the normalized time scale to the authors knowledge,
so a weighted mean has been used given by
*
*
Tcw* Tcurrent
U cw Twaves
(1 U cw )

(8)

The time scale of scour T can be calculated by

D2

g s 1d

3
50

T*

(9)

where g is the acceleration due to gravity, s is the relative density of the sediment, d50 is median grain
diameter.
By discretizing Eq.(1) and using the given equations (1) - (9), the scour evolution and backfilling has been
calculated for each time step of the environmental conditions. Seq in the model has been adopted for current
alone condition where Ucw>0.7 and the time scale of backfilling is taken equal to the time scale of scouring.
The model has been run by using the available hindcast data from October and November months. The time
step used for the prediction was the same as the resolution of the metocean data, namely 1 hour. Results
indicate that the backfilling in the model occurs too fast (Fig. 5). Field data has been used to tune the model.
The best fit has been obtained for the given period when the time scale of the backfilling is in order of 10
times the time scale of scouring (Fig. 6).

1445

ICSE6 Paris - August 27-31, 2012

F.H. Dixen et al.

Figure 5 : Comparison of field and model data.

Figure 6 : Tuned model data versus measurements

VI

CONCLUSIONS

The results of the field measurement enabled to capture the short time variations of scour development.
Following conclusions have been reached:
The maximum depth of scour observed during the measurement period is S/D=1.7. Considering the
time elapsed since installation, this value is considered as fully developed scour depth and it is found
slightly larger than the values found by using the existing formulae. [Breuser et.al,1977], [DNV,
2010].

1446

ICSE6 Paris - August 27-31, 2012

F.H. Dixen et al.

Scour depth increased despite of wave action and tidal reversal during the 6 months of measurement
period.
Variation in the scour depth followed the spring and neap tide.
Assuming the backfilling rate as 10 times the scouring rate, the time development of scour has been
predicted reasonably by using a relative simple mathematical model.

VII ACKNOWLEGMENTS AND THANKS

The research on which this paper is based was funded by DONG Energy Ltd. through the company
research programme. Support for hindcast study came from DHI and data acquisition and processing was
realized by EIVA.
VIII REFERENCES AND CITATIONS
Breusers, H.N.C., Nicollet, G. and Shen, H.W. (1977). Local scour around cylindrical piers. Journal of
Hydraulic Research, IAHR 15(3): 211-252.
DNV (Det Norske Veritas) (2010). - DNV-OS-J101.
Hoffmans, G.J.C.M., Verheij, H.J. (1997). - Scour Manual.
Sumer, B.M., Fredse, J. (2002). - The Mechanics of Scour in the Marine Environment.
Whitehouse R. (1998). - Scour at Marine Structures.

1447