The importance of marine wind farms, as artificial hard substrates, for the ecology of the ichthyofauna

peletonwhoopUrban and Civil

Nov 26, 2013 (3 years and 6 months ago)

81 views

The importance of marine wind farms, as artificial hard
1

substrates, for the ecology of the ichthyofauna

2


3

J.T. Reubens
1*
, S. Degraer
2,1
, M. Vincx
1

4


5

1
Ghent University, Department of Biology, Marine Biology Section, Krijgslaan 281 S8, 9000 Gent,
6

Belgium

7

2
Royal Belgian Institute of Natural Sciences, Management Unit of the North Sea Mathematical Models
8

(MUMM), Marine Ecosystem Management Section, Gulledelle 100, 1000 Brussels, Belgium

9

*

Corresponding author. Tel: +32 9 264 85 17; fax: +32 9 264 85 98; email:

Jan.Reubens@UGent.be

10


11


12


13

Scientific divers, relaxing after work






©Jan Reubens

14

15


16

Contents

17

Abstract

................................
................................
................................
................................
..................

3

18

1. Introduction

................................
................................
................................
................................
........

4

19

2. Material and Methods

................................
................................
................................
......................

5

20

2.1 St
udy site

................................
................................
................................
................................
.....

5

21

2.2 Data collection

................................
................................
................................
............................

5

22

2.2.1 Species richness, length
-
frequency and CPUE

................................
.............................

5

23

2.2.2 Pouting density estimation

................................
................................
................................

6

24

2.2.3 Stomach analyses

................................
................................
................................
..............

6

25

2.3 Data analysis

................................
................................
................................
..............................

7

26

3. Results

................................
................................
................................
................................
...............

7

27

3.1 Species richness, length
-
frequencies and

CPUE

................................
................................
.

7

28

3.2 Pouting densities

................................
................................
................................
........................

8

29

3.3 Stomach analyses of pouting

................................
................................
................................
...

9

30

4. Discussion

................................
................................
................................
................................
.........

9

31

4.1 Species richness, length
-
frequencies and CPUE

................................
................................
.

9

32

4.2 Pouting densities

................................
................................
................................
......................

10

33

4.3 Stom
ach analyses of pouting

................................
................................
................................
.

11

34

Acknowledgement

................................
................................
................................
..............................

12

35

References

................................
................................
................................
................................
..........

12

36


37

38


39

Abstract

40


41

A substantial expansion of offshore wind farms in the North Sea has been planned, inducing a growing
42

interest in the possible effects of these constructions on the marine environment. To date however,
43

little research has been done to consider the possible
effects on the ichthyofauna.

44

This study provided first insights in the use of the artificial hard substrates by
several fish species with
45

a focus on
Trisopterus luscus

(pouting) at the Thorntonbank wind farm (Belgian part of the North Sea).

46

Scuba diving
operated visual surveys, carried out between July and October 2009, revealed a
47

population size of at least 30 000 individuals of pouting, representing a biomass of 3.6 *
10
6

kg,

in the
48

vicinity of a wind turbine. Line fishing and gillnet fishing were conducted throughout 2009 to collect fish
49

for stomach content analysis. T
he results supported the prime importance of the hard substrate prey
50

species
Jassa herdmani

and
Pisidia longi
cornis

in the diet of pouting.

51

52


53

1. Introduction

54


55

Marine structures, whether natural or man
-
made, have the potential to attract and concentrate fishes
56

and/or to enhance local fish stocks
(Bohnsack, 1989; Pickering and Whitmarsh, 1997; Leitao

et al.
,
57

2008; 2009)
. Several mechanisms may stimulate this behaviour, including (1) shelter against currents
58

and predators
(Jessee

et al.
, 1985; Bohnsack, 1989)
, (2) additional food and favoured prey types
59

(Pike and Lindquist, 1994; Fabi

et al.
, 2006; Leitao

et al.
, 2007)
, (3) increased feeding efficiency and
60

(4) provision of nursery and recruitment sites
(Bull and Kendall Jr, 1994)
.

61

Whether these structures only attra
ct and concentrate fishes or also increase the local productivity is
62

however still subject to debate
(Bohnsack and Sutherland, 1985; Bohnsack, 1989; Polovina, 1989;
63

Pickering and
Whitmarsh, 1997)
. If fishes are merely attracted to the structures due to some
64

behavioural preferences, concentrating them at one site, the structure may act as an ecological trap
65

(Robertson and Hutto, 2006)
. In addition, the hard structures may promote

overfishing by increasing
66

the possible catch per unit effort (CPUE). If the structure however enhances the local productivity, the
67

environmental carrying capacity of the system increases as well
(Bohnsack, 1989)
.

68

The outcome of attra
ction versus production may differ in different locations and for different species
69

(Bohnsack and Sutherland, 1985)
.
For this reason it is important to interpret the dimensions and
70

distribution areas of the fish populations involv
ed and to determine factors influencing structure
71

(densities) and functionality (production versus dispersion) to quantify the ‘possible’ net production.

72

Some of the fish species observed in close proximity to artificial hard structures in the Belgian Part

of
73

the North Sea (BPNS) are
Trisopterus luscus

(Linnaeus, 1758) (pouting),
Gadus morhua

(Linnaeus,
74

1758) (cod),
Dicentrarchus labrax

(Linnaeus, 1758) (seabass),
Pollachius pollachius

(Linnaeus, 1758)
75

(pollack),
Trachurus trachurus

(Linnaeus, 1758) (horse
mackerel) and
Scomber scombrus

(Linnaeus,
76

1758) (mackerel)
(Zintzen

et al.
, 2006; Mallefet

et al
.
, 2007)
.

77

With the construction of a wind farm in the BPNS, initiated in 2008, a unique situation is offered to
78

investigate what mechanisms play a major role in attracting fishes to these artificial hard substrates. In
79

the present investigation general

information is gathered on fish species diversity, density of the fish
80

species present and length
-
frequency information. Moreover, an estimation of the abundance of
81

pouting in the vicinity of the wind farm was made

based on visual information
.
Additionally pouting
82

caught near the wind turbines was investigated to determine whether the epifauna present at the
83

foundations of these turbines is a key constituent of their diet.

84


85

2. Material and Methods

86

2.1 Study site

87

The C
-
Power wind farm under consi
deration is located at the Thorntonbank, a natural sandbank 27 km
88

offshore, in the BPNS. At present, six gravity
-
based foundations are built. In the near future a total of
89

54 wind turbines will be constructed on this sandbank, covering an area of approxima
tely 14 km
2
. By
90

2020 more than 200 wind turbines will be present in the BPNS.

91

Each foundation has a diameter of six metres at the surface expanding to 16 metres at the seabed,
92

which lies at a depth of 25 m at high tide. Each foundation is surrounded by a
scour protection layer,
93

consisting of two coats. The filter layer is made up by pebble ranging from 2.5 mm up to 75 mm and
94

has a diameter of 48 m (1800 m²). This layer is overtopped by the armour layer, consisting of a
95

protective stone mattress with rocks
ranging from 250 mm up to 750 mm and has a diameter of 44 m
96

(1600 m²).

97

The surrounding soft sediment seabed is composed of medium sand (mean median grain size 374
98

µm, SE 27 µm)(Reubens
et al
., 2009).

99


100

2.2 Data collection

101

2.2.1 Species richness, length
-
freq
uency and CPUE

102

Nine sampling campaigns were organised at the wind farm in 2009 (table 1). Line fishing was
103

conducted to collect

fish species
, since this is an efficient method to catch fish on hard substrates.
104

Hooks, size nr 4, from the brand Arca were use
d. Fresh or frozen lugworm (
Arenicola marina
) was
105

used as bait. Angling was performed 1 to 10 metres away from a turbine (i.e. within the erosion
106

protection layer radius) just above the bottom of the seabed, assuring catching individuals hovering at
107

the ar
tificial hard substrate. Which turbine (D1
-
D6) was sampled depended upon technical constraints.
108

It was assumed that no significant differences in species richness, density or length
-
frequency were
109

present between the turbines. Data were pooled and no dist
inction was made between turbines.

110

All fish species caught were identified, measured (total length) and weighed (wet weight). Catch per
111

unit effort (CPUE) was calculated for each species and for each sampling day. CPUE is a commonly
112

used index for relativ
e fish abundance measurements
(Haggarty and King, 2006)
, although it is often
113

considered to be biased and not necessarily proportional to abundance
(Harley

et al.
, 2001)
. This can
114

partly be con
trolled by the use of standardized fishing methods, duration and gear. In this research fish
115

bait, time duration, material and number of fishermen involved were standardized. In this way a
116

comparison in CPUE and relative abundances over time could be made.


117

2.2.2 Pouting density estimation

118

Nine fish surveys were carried out, on the scour protection of one wind turbine between July and
119

October 2009. Data were collected using underwater visual census performed by scuba divers. Prior
120

to the surveys a transect line was placed around the footing
of the wind turbine indicating nine
121

sampling points (A
-
I) (table 2). A variation of the stationary sampling method
(Bannerot and Bohnsack,
122

1986)
, using semi
-
circles, was applied. At each sampling point the average visibility was estimated by
123

tape measure
. This average visibility was used as radius for the semi
-
circle. Pouting was counted and
124

the size of the fish was noted. If too many fish were present for easy counting, a small portion of the
125

school was counted and an extrapolation was made to estimate t
he total density. Fish lengths were
126

estimated comparing the fishes to a ruler attached to a writing board.

127

2.2.3 Stomach analyses

128

Both line fishing and gill netting were conducted to collect pouting for stomach content analysis.
129

Sampling was performed thr
oughout the year 2009. Line fishing was performed since this is an
130

efficient method to catch fish on hard substrates. For line fishing the same sampling strategy as
131

mentioned in section 2.2.1 was used.

132

Gill nets were further used since this technique has
a much higher CPUE than angling. A commercial
133

cod net (270 m in length and 1 m high), with a mesh size of 50 mm was used. The net was set for 2 up
134

to 4 hours.

135

After being measured (total length) and weighed (wet weight), 94 specimens of pouting were gutted
.
136

Their digestive system was preserved in an 8% formaldehyde
-
seawater solution. All food components
137

in the digestive tracts were identified to the lowest possible taxonomic level. Dry weight (60 °C for 48
138

h) and ash weight (500 °C for 2 h) were measured fo
r all food components.

139


140

2.3 Data analysis

141

To assess dietary composition an occurrence (%FO) and abundance (%A) method were used
142

(Hyslop, 1980)
. The relative abundance (%A
i
) can be either numerical or gravimetrical. For the
143

gravimetrical analysis ash
-
free dry weight (AFDW) was used.

144

%FO
i

= (N
i
/N)*100

145







%A
i

= (∑S
i
/∑S
a
)*100

146

N
i

is the number of predators with prey type
i

in their stomach, N the total number of no
-
empty
147

stomachs, S
i

is the stomach content composed by prey
i

and S
a

the total stomach content of all
148

stomachs together
(Amundsen

et al.
, 1996)
.

149

The feeding coefficient (Q)
(Hureau, 1970)

and the index of relative importance(IRI)
(Pinkas

et al.
,
150

1971)

were used to evaluate the dietary importance of each food category.

151

The Statistica software package was used to execute the Analyses of Variance (ANOVA) and the non
-
152

parametric tests.
The PRIMER v6 software package
(Clarke and Gorley, 2006)

was used to run
153

Principal Component Analysis (PCA).
For the data matrices a dis
tinction was made between numerical
154

and gravimetrical information. For both datasets relative abundances of prey were used instead of the
155

rough data, to detect patterns in individual diet and foraging behaviour
(De Crespin de Billy

et al.
,
156

2000)
. The relative abundance (%R
i
) of prey type
i

is described by the equation:

157

%R
i
= (S
i
/S
t
)*100

158

in which S
i

is the stomach content (number or weight) composed by prey
i
, and S
t

the total sto
mach
159

content of the individual predator (based on Amundsen
et al.

(1996)
). Prey types that appear in only
160

one stomach, having a low representativeness, were excluded f
rom the multivariate analyses.

161

3. Results

162

3.1 Species richness, length
-
frequencies and CPUE

163

Over all sampling campaigns 7 different fish species were caught: cod, pouting, mackerel, horse
164

mackerel, saithe (
Pollachius virens
; Linnaeus, 1758), black seabream

(
Spondyliosoma cantharus
;
165

Linnaeus, 1758) and bull rout (
Myoxocephalus scorpius
; Linnaeus, 1758). Saithe, black seabream and
166

bull rout
were
caught only once.

These rare species were not further analysed.

Mackerel and horse
167

mackerel were caught in summer a
nd early autumn, while cod and pouting
(figure 1 and 2)
were
168

caught during most of the sampling campaigns.
Length of mackerel varied between 24 and 31 cm
169

(average 27.20 cm; SD 1.89 cm).
No sig
nificant differences in length we
re present between the
170

differen
t months (Mann
-
Withney U Test, p > 0.05).

The length of horse mackerel ranged from 21 to 29
171

cm (average 23.21 cm; SD 2.52 cm). A significant difference in length was present between July and
172

September (One
-
way Anova, p = 0.016). Cod lengths varied between

20 and 57 cm (average 36.31
173

cm; SD 7.95 cm). Length differed significant between seasons (one
-
way Anova, P < 0.05). The
174

average length is lowest in February (23.82 cm) and highest in October (41.71 cm) (figure 1). The
175

length of pouting ranged from 13 to
34 cm (average 23.03 cm; SD 3.52 cm). In summer and early
176

autumn a broader range in length classes was present than in winter (figure 2). Length differed
177

significantly between months, but no clear seasonal pattern in length classes was present.

178

Catch per u
nit effort (CPUE) data was recorded to get an idea of the abundances of fishes present in
179

the vicinity of the wind turbines.

Total CPUE ranged between 0 and 28. In winter and early spring
180

CPUE was much lower (0
-

5) than in summer and autumn (8


28), alth
ough the period of highest
181

CPUE is species dependent. For cod CPUE is higher in winter than in summer, while mackerel and
182

horse mackerel don’t appear in winter.

Pouting was present all through the year. For the latter species
183

CPUE was much higher in summer

and autumn (maximum CPUE of 15.6) than in winter and early
184

spring (maximum CPUE of 2).

185

The ecology of pouting was investigated in more detail. Density estimations were made based on
186

visual observations and feeding behaviour in the vicinity of the wind tur
bines was investigated.

187


188

3.2 Pouting densities

189

The visual surveys performed highlight the value of the artificial hard substrates for pouting. This
190

species was present in high abundances near the wind turbine during all surveys (7
-
74 specimens/m
2
)
191

(table
2). To estimate the population size it was stated that the fish were evenly distributed across the
192

erosion protection layer (1600 m
2
).

193


A mean local population of 30 000 individuals (mean length 20 cm) was assessed near one wind
194

turbine. A large variation

in numbers (10 000
-

118 000) was observed over time. Based on a Length
-

195

Wet weight relationship
(Merayo and Villegas, 1994)
, the population had a minimum biomass of

3.6 *
196

10
6

kg. Similar abundances of pouting may be present near the other wind turbines as mean CPUE of
197

comparable magnitudes (spring 0
-

2.5; summer 8
-

15.5) were recorded throughout the year. Once
198

the wind farm will accomplish its full capacity it may
harbour a population of pouting with a total
199

biomass reaching 216 * 10
6

kg.

200

3.3 Stomach analyses of pouting

201

Fish caught by line and gill net fishing weighed 95 g up to 195 g and lengths varied between 18.4 cm
202

and 24.2 cm. Of the 94 stomachs analysed, 12 we
re empty. The diet of
T. luscus

contained a wide
203

variety of food items: 46 prey types were identified, although 20 occurred only once in the stomachs
204

analysed. Of the 46 prey types 12 are hard substrate associated, while 9 are restricted to soft
205

sediments

(Table
3
).
Jassa herdmani
,
Jassa
mats,
Pisidia longicornis
, Brachyura sp. and detritus
206

were the prey types with the highest %FO.
Liocarcinus holsatus
, fish scales,
Phtisica marina
,
207

Nematoda sp. and
Mytilus edulis

were also frequently present in the stoma
chs.

208

Jassa herdmani

and
P. longicornis

were the only prey species composing more than 10% of the total
209

numerical prey abundance. For the gravimetrical measures
P. longicornis
,
J. herdmani

and
Jassa

210

mats composed more than 10% of the total AFDW (table
3
).

Q and IRI indicated that
J. herdmani

and
211

P. longicornis

were the most important prey species contributing to the diet of pouting. Both species
212

are restricted to hard substrates. Some prey types however could not be quantified, preventing Q and
213

IRI to be calculated.

214

Pisidia longicornis
,
J. herdmani
, detritus a
nd
Jassa

mats were the most differentiating prey types
215

(variables) in the diet of pouting

based on gravimetrical information
(fig 1.). Numerically,
J. herdmani

216

and
P. longicornis

dominated the gut contents (fig. 2). Differentiating as well, but to a lesser
extent
217

were Nematoda sp. and Pisces sp. In both the gravimetrical and numerical analysis many

samples
218

were

positioned at the edge of one of the explaining variables, demonstrating high selectivity for the
219

particular prey and a weak within
-
individual variat
ion. The samples positioned near the origin or amid
220

the explanatory variables expressed less selectivity for prey and/or foraging on rare prey species.

221


222

4. Discussion

223

4.1 Species richness, length
-
frequencies and CPUE

224

The fish species observed in the vicinity of the wind farm are in agreement with the species expected
225

to be present
(Farm, 2006; Zintzen

et al.
, 2006; Mallefet

et al.
,

2007)

The same species are frequently
226

observed near other types of hard structures, like shipwrecks, in the BPNS. Although being attracted
227

to hard substrates, seabass was the single species expected t
hat was not observed near the wind
228

turbines. This species however is very bashful and sensitive to noise (pers. comm.), which might
229

explain why seabass wasn’t observed nor caught.

230

Mackerel was only present during summer and early autumn. In winter this spe
cies is present in
231

deeper water. Accordingly to their length, which ranged from 24 to 31 cm, the mackerel are between 1
232

and 2 years old (ICES, 2006). Consequently, most individuals are mature. Horse mackerel were only
233

present during summer and early autumn
, which is in agreement with Vandendriessche
et al.

(2009).
234

Mean length however, is much higher for horse mackerel hovering near the hard substrate in
235

comparison with individuals on the soft sediment (Vandendriessche
et al.

2009). Consequently older
236

indivi
duals are present in the vicinity of the hard substrates. For cod, age ranges from 1 to 4 years old
237

(Heessen, 1983)
. In January and February most fish are between 1 and 2 years old, while in summer
238

and autumn most species are between 2 and 4 years old. Some cod mature in their second year of
239

life, but it is not before the age of six that al
l are mature (ICES, 2006). Consequently, most cod present
240

are juveniles, which is in agreement with the observations at the wind farm Horns rev
(Farm, 2006)
.
241

Accordingly to their length, which ranged from 13 to 34 cm, the pouting are between 0 and 4 yea
rs old
242

(Merayo and Villegas, 1994)
. The majority of the fish present are between 1 and 3 years old.
243

Compared to pouting present on the soft sediment in the surrounding

area, larger and older fish are
244

present near the hard substrates (Vandendriessche
et al.

2009).

245

No information is available from CPUE data on other hard substrates in the BPNS. Compared to the
246

soft sediments (Vandendriessche
et al
; 2009)
densities are highly enhanced
for pouting and horse
247

mackerel
near the artificial hard substrates of the wind turbines, indicating the aggregation effect of the
248

turbines.
For mackerel and cod I’m awaiting information of ILVO
. It should be noted however that

249

density estimation on the hard and soft substrates are gathered in a different way, which may
250

influence CPUE data. Near the hard substrates line fishing was performed, while on the soft sediments
251

beam trawl was used to catch fish.

252


253

4.2 Pouting densities

254

Previous research has illustrated that artificial hard substrates in many parts of the world, attract
255

several fish species
(Petersen and Malm, 2006; Wilhelmsson

et al.
, 2006)
. In the BPNS pouting is
256

frequently observed at artificial hard struct
ures
(Zintzen

et al.
, 2006; Mallefet

et al.
, 2007)

which is
257

consistent with the results obtai
ned by the current research. Visual observations by scuba diving
258

indicated that a large local population of pouting (30 000 individuals) was present in the vicinity of the
259

wind turbine investigated. Once the wind farm will have its full capacity (60 wind t
urbines) this
260

represents a minimum biomass of 216 * 10
6

kg. In comparison, between 2000 and 2006 roughly 400
261

to 500 * 10
6

kg of pouting were landed in the Belgian harbours annually (Fishstat Plus, FAO 2008).

262

It is interesting to note however, that the pop
ulation size near the wind turbines was actually higher
263

than estimated. (1)Visual census methods underestimate abundant fish species
(Sale and Douglas,
264

1981; Brock, 1982; Bannerot and Bohnsack, 1986)
. (2) Although pouting is observed too in high
265

densities near the foundation, the estimation was restricted to the erosion protection layer, as
266

abundances near the fo
rmer are more difficult to estimate. (3)

Using a stationary observation method
267

in low visibility waters induces an extra source of underestimation. Individuals located at the outer
268

edges of the visibility range are more difficult to detect and often overlo
oked. No literature is available
269

for densities of pouting present on other artificial hard substrates in the BPNS. The population size is
270

hence to be considered a minimum estimate.

271

Pouting densities are highly enhanced near the artificial hard substrates
(7
-
74 specimens/m², based
272

on visual observations) of the wind turbines in comparison with those on the soft sediments (< 0.001
273

specimens/m², based on beam trawl data) in the surrounding area of the wind farm (Vandendriessche
274

et al
. 2009), indicating the ag
gregation effect of the turbines.

275

4.3 Stomach analyses of pouting

276

Stomach content analysis was performed on 94 pouting. Indicated by the weights and lengths the fish
277

belonged to year class 1 and 2
(Merayo and Villegas, 1994; Heessen and Daan, 1996)
.

278

Based on the overall outcome of the analyses
J. herdmani
,
P. longicornis

and
Jassa

mats were found
279

to be the most abundant and differentiating prey types in the diet of pouting. It is interesting to note
280

that
J. herdmani

is a tube
-
dwelling amphipod. The tube
-
mats themselves (here called
Jassa
mats) can
281

be indicated as a particular kind
of detritus. These tube
-
mats were ingested in high quantities (table
3
).
282

Visual observations however, revealed that pouting does not pick the individual
Jassa

specimens out
283

of the tube mats, but bite off a mouthful of mat together with
Jassa
. Subsequently
it is assumed that
284

the mats themselves are of lesser importance in the diet of pouting.

285

It should be noted that both
J. herdmani

and
P. longicornis

are established in very high abundances
286

on the wind turbine foundations (Kerckhof
et al.

2009). In other wind farms in the North Sea and on
287

shipwrecks these prey are also frequently present in high abundances
(Schröder

et al.
, 2006; Mallefet

288

et al.
, 2007)
. The foundations of the wind turbines are densely colonized and have high species
289

diversity (Kerckhof et

al. 2009), indicating food is plentiful for many predators. Many prey types
290

present in the diet of pouting (table
3
) are established on the wind turbines (Kerckhof
et al.

2009).

291


292

Our results suggest that pouting benefits from the artificial hard substrate
s of the wind farm. Pout
293

densities are highly enhanced in comparison

with those on the soft sediments. Food is plentiful on the
294

hard substrates and many prey types found are also present on the turbines as epifauna. The
295

dominant food items
J. herdmani

and
P. longicornis

in the diet of pouting are established in very high
296

densities on the turbine foundations.

297


298

Acknowledgement

299

The first author acknowledges a FWO predoctoral grant. This research was facilitated by the Flanders
300

Marine Institute (VLIZ). We are t
hankful to the crew of the RV “Zeeleeuw”, the diving team and the
301

numerous colleagues for their assistance in the field. We thank the VLIZ and the Management Unit of
302

the North Sea Mathematical Models (MUMM) for their technical support. The responsibility f
or any
303

remaining errors remains with the author.

304


305

References

306

Amundsen, P. A., Gabler, H. M., and Staldvik, F. J. 1996. A new approach to graphical analysis of
307

feeding strategy from stomach contents data

modification of the Costello
method. Journal of
308

Fish Biology, 48: 607
-
614.

309

Bannerot, S. P., and Bohnsack, J. A. 1986. A stationary visual census technique for quantitatively
310

assessing community structure of coral reef fishes. NOAA Technical Report NMFS, 41: 1
-
15.

311

Bohnsack, J. A. 1989.

Are high densities of fishes at artificial reefs the result of habitat limitation or
312

behavioral preference? Bulletin of Marine Science, 44: 631
-
645.

313

Bohnsack, J. A., and Sutherland, D. L. 1985. Artificial reef research: a review with recommendations
314

for f
uture priorities. Bulletin of Marine Science, 37: 11
-
39.

315

Brock, R. E. 1982. A critique of the visual census method for assessing coral reef fish populations.
316

Bulletin of Marine Science, 32: 269
-
276.

317

Bull, S., and Kendall Jr, J. J. 1994. An indication of th
e process: offshore platforms as artificial reefs in
318

the Gulf of Mexico. Bulletin of Marine Science, 55, 2: 1086
-
1098.

319

Clarke, K. R., and Gorley, R. N. 2006. PRIMER v6: user manual/tutorial PRIMER
-
E. Plymouth, UK.

320

De Crespin de Billy, V., Doledec, S., and
Chessel, D. 2000. Biplot presentation of diet composition
321

data: an alternative for fish stomach contents analysis. Journal of Fish Biology, 56: 961
-
973.

322

Fabi, G., Manoukian, S., and Spagnolo, A. 2006. Feeding behavior of three common fishes at an
323

artificia
l reef in the northern Adriatic Sea. Bulletin of Marine Science, 78: 39
-
56.

324

Farm, H. 2006. Hydroacoustic Monitoring of Fish Communities in Offshore Wind Farms.

Horns Rev
325

Offshore Wind Farm
-

Annual Report, 54 pp

326

Haggarty, D. R., and King, J. R. 2006. CPUE
as an index of relative abundance for nearshore reef
327

fishes. Fisheries Research, 81: 89
-
93.

328

Harley, S. J., Myers, R. A., and Dunn, A. 2001. Is catch
-
per
-
unit
-
effort proportional to abundance?
329

Canadian Journal of Fisheries and Aquatic Sciences, 58: 1760
-
177
2.

330

Heessen, H. J. L. 1983. Distribution and abundance of young cod and whiting in the south
-
eastern
331

North Sea in the period 1980

1982.
ICES Document CM.

332

Heessen, H. J. L., and Daan, N. 1996.
Long
-
term trends in ten non
-
target North Sea fish species. Ices
333

Journal of Marine Science, 53: 1063.

334

Hureau, J. C. 1970. Biologie comparée de quelques poissons antarctiques (Nototheniidae). Bull. Inst.
335

Océanogr. Monaco, 68: 1
-
224.

336

Hyslop, E. J. 1980. Stomach contents analysis
-

a review of methods and their application
. Journal of
337

Fish Biology, 17: 411
-
429.

338

ICES, 2006.
http://www.ices.dk/marineworld/ices
-
fishmap.asp

339

Jessee, W. N., Carpenter, A. L., and Carter, J. W. 1985. Distribution patterns and density estimates of
340

fishes on a southern California artificial reef with

comparisons to natural kelp
-
reef habitats.
341

Bulletin of Marine Science, 37: 214
-
226.

342

Kerckhof, F., Norro, A., Jacques, T., Degraer, S., 2009. Early colonisation of a concrete offshore
343

windmill foundation by marine biofouling on the Thornton Bank (southern
North Sea).
In

344

Offshore wind farms in the Belgian part of the North Sea: State of the art after two years of
345

environmental monitoring. pp. 39
-
51.

Ed. by S. Degraer and R. Brabant. Brussels. 287 pp. +
346

annexes

347

Leitao, F., Santos, M. N., Erzini, K., and Monte
iro, C. C. 2008. Fish assemblages and rapid colonization
348

after enlargement of an artificial reef off the Algarve coast (Southern Portugal). Marine
349

Ecology
-
an Evolutionary Perspective, 29: 435
-
448.

350

Leitao, F., Santos, M. N., Erzini, K., and Monteiro, C. C.
2009. Diplodus spp. assemblages on artificial
351

reefs: importance for near shore fisheries. Fisheries Management and Ecology, 16: 88
-
99.

352

Leitao, F., Santos, M. N., and Monteiro, C. C. 2007. Short communication, Contribution of artificial
353

reefs to the diet of

the white sea bream (Diplodus sargus). Ices Journal of Marine Science, 64.

354

Mallefet, J., Zintzen, V., Massin, C., Norro, A., vincx, M., De Maersschalck, V., Steyaert, M., et al. 2007.
355

Belgian shipwreck: hotspots for marine biodiversity (BEWREMABI). Belgia
n Science Policy
356

Office, Brussels. 155 pp.

357

Merayo, C. R., and Villegas, M. L. 1994. Age and growth of
Trisopterus luscus

(Linnaeus, 1758) (Pisces,
358

Gadidae) off the coast of Asturias. Hydrobiologia, 281: 115
-
122.

359

Petersen, J. K., and Malm, T. 2006. Offshore

windmill farms: threats to or possibilities for the marine
360

environment. AMBIO: A Journal of the Human Environment, 35: 75
-
80.

361

Pickering, H., and Whitmarsh, D. 1997. Artificial reefs and fisheries exploitation: a review of the
362

‘attraction versus production
’debate, the influence of design and its significance for policy.
363

Fisheries Research, 31: 39
-
59.

364

Pike, L. A., and Lindquist, D. G. 1994. Feeding ecology of spottail pinfish (Diplodus holbrooki) from an
365

artificial and natural reef in Onslow Bay, North Carol
ina. Bulletin of Marine Science, 55, 2: 363
-
366

374.

367

Pinkas, L., Oliphant, M. S., and Iverson, I. L. 1971. Food habits of albacore, bluefin tuna, and bonito in
368

California waters. Fish Bull., 152: 1
-
105.

369

Polovina, J. J. 1989. Artificial reefs: nothing more than

benthic fish aggregators. Reports of California
370

Cooperative Oceanic Fisheries Investigations, 30: 37
-
39.

371

Reubens, J., Vanden Eede, S., Vincx, M. 2009.
Monitoring of the effects of offshore wind farms on the
372

endobenthos of soft substrates: Year
-
0 Bligh Ban
k and Year
-
1 Thorntonbank.
In

Offshore wind
373

farms in the Belgian part of the North Sea: State of the art after two years of environmental
374

monitoring. pp. 61
-
91. Ed. by S. Degraer and R. Brabant. Brussels. 287 pp. + annexes

375

Robertson, B. A., and Hutto, R. L
. 2006. A framework for understanding ecological traps and an
376

evaluation of existing evidence. Ecology, 87: 1075
-
1085.

377

Sale, P. F., and Douglas, W. A. 1981. Precision and accuracy of visual census technique for fish
378

assemblages on coral patch reefs. Enviro
nmental Biology of Fishes, 6: 333
-
339.

379

Schröder, A., Orejas, C., and Joschko, T. (2006). Benthos in the vicinity of the piles: FINO 1 (North Sea)
380

In
Offshore Wind Energy. Research on Environmental Impacts

pp. 185
-
200. Ed. by J. Köller, J.
381

Köppel &

W. Peters.
Springer Verlag Heidelberg, Berlin. 371 pp.

382

Vandendriessche, S.; Hostens, K.; Wittoeck, J., 2009.
Monitoring of the effects of the Thorntonbank
383

and Bligh Bank windmill parks on the epifauna and demersal fish fauna of soft
-
bottom
384

sediments: Thor
ntonbank: status during construction (T1) Bligh Bank: reference condition (T0),
385

in
: Degraer, S.; Brabant, R. (Ed.) (2009).
In

Offshore wind farms in the Belgian part of the North
386

Sea: State of the art after two years of environmental monitoring. pp. 93
-
150
.
Ed. by S.
387

Degraer and R. Brabant. Brussels. 287 pp. + annexes

388

Wilhelmsson, D., Malm, T., and Ohman, M. C. 2006.
The influence of offshore windpower on
389

demersal fish. Ices Journal of Marine Science, 63: 775.

390

Zintzen, V., Massin, C., Norro, A., and Mallefe
t, J. 2006. Epifaunal inventory of two shipwrecks from
391

the Belgian Continental Shelf. Hydrobiologia, 555: 207
-
219.

392


393

394


395

Table 1. Overview of the nine angling campaigns performed in the wind farm.

396

Date

7/01/2009

20/01/2009

3/02/2009

4/02/2009

4/03/2009

2/07/2009

29/09/2009

27/10/2009

6/11/2009

Location

D3

D4

D4

D2, D4

D2

D3, D4

D4, D5

D6

D4

Species

T. luscus

G. morhua

T. luscus

G. morhua

T. luscus

G. morhua

T. luscus

G. morhua

P. virens

T. luscus

T. luscus

G. morhua

T. trachurus

S. scumbrus

T. luscus

G. morhua

T. minutes

T. trachurus

S. scumbrus

T. luscus

G. morhua

S. scumbrus

S. canthurus

T. luscus












397


398

Table 2.Overview of the nine visual surveys performed to estimate pouting density.

399

Period

Jul
-
09

Jul
-
09

Jul
-
09

Aug
-
09

Sep
-
09

Sep
-
09

Sep
-
09

Oct
-
09

Oct
-
09

Sampling point

A

G

F

B

E

I

F

E

F

Mean observed numbers

140

93

435

150

127

167

150

150

100

Visibility range (m)

3

3

5.1

3.7

2

1.2

3.4

3

2

Individuals/m²

10

7

11

7

20

74

8

11

16












400


401

Length-frequency distribution of cod per month in 2009
Total length (cm)
No of obs
January
20
25
30
35
40
45
50
55
60
0
1
2
3
4
5
6
7
February
20
25
30
35
40
45
50
55
60
March
20
25
30
35
40
45
50
55
60
April
20
25
30
35
40
45
50
55
60
May
20
25
30
35
40
45
50
55
60
0
1
2
3
4
5
6
7
June
20
25
30
35
40
45
50
55
60
July
20
25
30
35
40
45
50
55
60
August
20
25
30
35
40
45
50
55
60
September
20
25
30
35
40
45
50
55
60
0
1
2
3
4
5
6
7
October
20
25
30
35
40
45
50
55
60
November
20
25
30
35
40
45
50
55
60
December
20
25
30
35
40
45
50
55
60
No data available
No data available
No data available
No data available
No data available

Figure 1. Length
-
frequency distribution of
cod

per month in 2009.


Length-frequency distribution of pouting per month in 2009
Total length (cm)
No of obs
January
12
14
16
18
20
22
24
26
28
30
32
34
0
2
4
6
8
10
12
14
February
12
14
16
18
20
22
24
26
28
30
32
34
March
12
14
16
18
20
22
24
26
28
30
32
34
April
12
14
16
18
20
22
24
26
28
30
32
34
May
12
14
16
18
20
22
24
26
28
30
32
34
0
2
4
6
8
10
12
14
June
12
14
16
18
20
22
24
26
28
30
32
34
July
12
14
16
18
20
22
24
26
28
30
32
34
August
12
14
16
18
20
22
24
26
28
30
32
34
September
12
14
16
18
20
22
24
26
28
30
32
34
0
2
4
6
8
10
12
14
October
12
14
16
18
20
22
24
26
28
30
32
34
November
12
14
16
18
20
22
24
26
28
30
32
34
December
12
14
16
18
20
22
24
26
28
30
32
34
No data available
No data available
No data available
No data available
No data available

Figure 2. Length
-
frequency distribution of
pouting
per month in 2009
.


Table 3. List of prey items present in the
stomachs of pouting (
Trisopterus luscus
).. Frequency of occurrence (%FO), densities
(%dens), ash
-
fry dry weight (%AFDW), feeding coefficient (Q) and index of relative importance (IRI).
N/A indicates that no
quantification could be made or the information i
s missing.
H

Taxa living on hard substrates.
S

Taxa living on soft substrates.
B

Taxa found on both substrates.
N/A

Not applicable.


SPECIES

%FO

% dens

% AFDW

Q

IRI


Hydrozoa






H

Unidentified sp.

3.66

0.03

0.01

<0.01

0.17

H

Halecium
sp.

1.22

N/A

0.01

N/A

N/A


Nematoda






N/A

Unidentified sp.

9.76

0.57

0.02

<0.01

5.71


Polychaeta






N/A

Unidentified sp.

8.54

0.17

1.36

0.23

13.00


Crustacea






N/A

Unidentified sp.

3.66

0.03

0.04

<0.01

0.27


Cirripedia






H

Unidentified sp.

8.54

0.23

0.03

<0.01

2.23

H

Balanidae sp.

3.66

0.17

0.37

0.06

1.98


Mysidasea






S

Acanthomysis longicornis

1.22

0.03

<0.01

<0.01

0.05

S

Gastrosaccus spinifer

3.66

0.17

0.02

<0.01

0.67


Amphipoda






N/A

Unidentified sp.

1.22

0.03

<0.01

<0.01

0.04

B

Amphilochus neapolitanus

1.22

0.07

<0.01

<0.01

0.09

B

Stenothoe marina

1.22

0.03

<0.01

<0.01

0.04

B

Corophium
sp.

1.22

0.03

<0.01

<0.01

0.04

H

Jassa herdmani

64.63

84.07

18.48

1553.26

6627.86

H

Jassa

mats

67.07

N/A

10.24

N/A

N/A

H

Caprella

sp.

1.22

0.03

<0.01

<0.01

0.04

H

Phtisica marina

9.76

0.84

0.03

0.02

8.42

S

Megaluropus agilis

1.22

0.03

<0.01

<0.01

0.04


Decapoda






N/A

Unidentified sp.

3.66

0.10

0.17

0.02

1.00


Natantia






N/A

Unidentified sp.

6.10

0.17

0.29

0.05

2.79

S

Processa
edulis crassipes

1.22

0.13

0.31

0.04

0.54

S

Processa modica

1.22

0.03

0.10

<0.01

0.17

N/A

Crangonidae

sp
.

1.22

0.03

0.07

<0.01

0.13

S

Crangon crangon

1.22

0.03

0.05

<0.01

0.10


Reptantia






N/A

Unidentified sp.

2.44

0.07

0.14

0.01

0.50

B

Paguridae
sp.

2.44

0.17

0.80

0.13

2.36

B

Pagurus bernhardus

2.44

0.10

1.39

0.14

3.64

N/A

Brachyura
sp.

12.20

0.37

1.23

0.45

19.48

H

Pisidia longicornis

35.37

10.45

46.55

486.69

2016.16

H

Macropodia linaresi

1.22

0.03

0.01

<0.01

0.05

S

Corystes cassivelaunus

1.22

0.03

0.09

<0.01

0.15

B

Portunidae sp.

2.44

0.13

1.57

0.21

4.15

B

Liocarcinus

sp.

1.22

0.03

2.07

0.07

2.57

B

Liocarcinus holsatus

9.76

0.47

5.22

2.44

55.46

B

Carcinus maenas

1.22

0.03

0.14

<0.01

0.21


Bivalvia






N/A

Unidentified sp.

1.22

0.03

0.07

<0.01

0.13

H

Mytilus edulis

9.76

0.77

0.02

0.02

7.65


Bryozoa






H

Unidentified sp.

3.66

N/A

1.36

N/A

N/A


Echinodermata






H

Asterias rubens

1.22

0.03

0.01

<0.01

0.06

N/A

Echinoidea sp.

1.22

0.03

<0.01

<0.01

0.04


Pisces






N/A

Unidentified sp.

4.88

0.13

2.01

0.27

10.47

S

Callionymus lyra

2.44

0.07

0.37

0.03

1.07

S

Callionymus reticulatus

1.22

N/A

1.72

N/A

N/A


Others






N/A

Detritus

10.98

N/A

2.66

N/A

N/A

N/A

Plant material

2.44

N/A

0.26

N/A

N/A

N/A

Fish scales

9.76

N/A

0.08

N/A

N/A