IC E S Journal o f M arine Science, 59: 1199-1214. 2002
doi:10.1006/jm sc.2002.1288, available online a t http://w w w .idealibrary.com on
^
Diversity and community structure o f epibenthic invertebrates
and fish in the North Sea
R. Callaway, J. Alsvâg, I. de Boois, J. Cotter, A. Ford,
H. Hinz, S. Jennings, I. Kröncke, J. Lancaster, G. Piet,
P. Prince, and S. Ehrich
C allaw ay, R ., Alsvâg, J., d e Boois, I., C o tte r, J., F o rd , A ., H inz, H ., Jennings, S.,
K rö n c k e, I., Lancaster, J., Piet, G ., Prince, P., a n d E hrich, S. 2002. D iversity and
co m m u n ity structure o f epibenthic in v erteb rates and fish in th e N o rth Sea. - IC E S
J o u rn a l o f M arine Science, 59: 1199-1214.
T h e stru c tu re o f N o rth Sea b enthic in v erteb rate and fish com m unities is an im p o rta n t
in d ic a to r o f anthropogenic a n d en v iro n m en ta l im pacts. A lth o u g h N o rth Sea fish
sto c k s are m onitored regularly, b enthic fa u n a are not. H ere, w e re p o rt th e results o f a
survey carried o u t in 2000, in w hich five n a tio n s sam pled th e epibenthic a n d fish fauna
a t 270 statio n s th ro u g h o u t th e N o rth Sea. T h e aim o f th e survey w as to investigate the
div ersity a n d com m unity stru c tu re o f epib en th ic and fish com m unities a n d to identify
relatio n sh ip s with environm ental factors, including th e frequency o f com m ercial o tte r
a n d beam traw ling disturbance. E pibenthic species diversity w as low er in th e so u th ern
N o rth Sea th a n in central a n d n o rth e rn areas. F ish, conversely, were m o re diverse in
th e so u th . T he 50 m , 100 m a n d 200 m d e p th c o n to u rs b ro a d ly defined the b o u n d aries
o f b e n th ic a n d fish com m unities. T h e a b u n d an c e o f e p ib e n th o s o f th e so u th e rn N o rth
S ea w as dom inated by free-living species, w hilst n o rth o f th e 50 m c o n to u r sessile
species prevailed. A hybrid a rea , w ith sessile species typical o f th e n o rth a n d free-living
species characteristic o f th e so u th , w as fo u n d off th e N o rfo lk and F lam b o ro u g h coast
stre tc h in g tow ards the D ogger Bank.
L arge-scale hydrodynam ic p h e n o m e n a w ere m ost likely to be responsible fo r the
m ain divisions betw een com m unities, especially the b o u n d a ry betw een m ixed a n d
stratified w ater m asses. H ow ever, b o tto m tem p e ra tu re, sedim ent p a ram ete rs a n d beam
traw lin g were closely co rrelated w ith species richness a n d diversity, as well as
co m m u n ity pattern s, and m ay m odify reg io n al species com position.
O u r study show s th a t effective large-scale sam pling o f b e n th ic com m unities can be
c o n d u cted du rin g existing fisheries surveys. Since an n u al fisheries surveys a re c o n ­
d u c te d th ro u g h o u t the n o rth e a st A tlan tic sh e lf seas, co n cu rre n t benthic surveys w ould
allow benthic sam pling on u n p recedented sp atial and tem p o ral scales. T he sam ples
w o u ld help to m o n ito r the en v iro n m en tal im pacts o f traw lin g d isturbance, clim ate
ch an g e, pollu tio n and o th er n a tu ra l a n d a n th ro p o g e n ic factors.
© 2002 International Council for the Exploration o f the Sea. Published by Elsevier Science Ltd.
All rights reserved.
K eyw ords: epibenthos, fish, diversity, co m m u n ity structure, fishing effects, N o rth Sea.
R eceived 11 F ebruary 2002; accepted 8 A pril 2002.
R. Callaway (née R. Z ü h lk e), J. Lancaster, a n d A. Ford: School o f B iological Sciences,
U niversity o f Wales Swansea, Singleton Park, Swansea, S A 4 3 S W , UK. H H in z and
I. K röncke: Forschungsinstitut Senckenberg, Schleusenstrasse 39a, D -26382
W ilhelmshaven, Germany. J. Alsvâg: In stitu te f o r M arine Research, Bergen, Norway.
I. D e Boois and G. Piet: Netherlands In stitu te f o r Fisheries Research ( R IV O ), P O B o x
58, 1970 A B Ijmuiden, The Netherlands. P. Prince: D anish Institute f o r Fisheries
Research, Hirtshals, Denmark. J. C otter a n d S. Jennings: Centre f o r Environm ent,
Fisheries and Aquaculture Science, L o w e sto ft Laboratory, N R 3 3 0H T , UK. S. Ehrich:
Bundesforschung-Sanstalt fü r Fischerei, In stitu t f ü r Seefischerei, Palm aille 9, D -22767
H am burg, Germany. Correspondence to R. Callaway: e-mail: bdzuhlke@ sw ansea.ac.uk
Introduction
Several studies have described th e diversity and co m m u ­
n ity stru ctu re o f epibenthic invertebrates an d dem ersal
1054-3139/02/061199+16 S35.00/0
\ ,., .
'■
(V7Mf)
\r
„
'
'
•"L A ÍV D F R * -'
Öös f
'
>■-j- t , —.
°
IN S T IT U
' ' - ° 1C n Q 'e - L' üj g i un - ,
fish in the N o rth Sea (Petersen, 1914; D y er et al., 1983;
B asford et a i , 1989; F rauenheim et al., 1989; K röncke,
1990; D uineveld et al, 1991; R o g ers et al., 1998; Rees
et al., 1999). H ow ever, all the studies o f epibenthos used
© 2002 In te rn a tio n a l C ouncil f o r th e E x p lo ra tio n o f th e Sea. P ublished by Elsevier S cience L td . A ll rig h ts reserved.
R. Callaway et al.
1200
E6
ICES rectangles
FO
F2
F4
ES
F6
F8
6 2 °N
mmm
CG
g
A N orw ay
O T h e N e th e rla n d s
▼ G erm an y
□
E ngland
D en m a rk
4°W
2°
0°
2 CE
4°
6°
8°
L o n g itu d e
Figure 1. S am pling stations.
different sam pling m ethods an d th e analyses w ere often
based o n a lim ited n um ber o f sta tio n s. F ro m 1996 to
1998 a E u ropean project sta n d ard ized th e sam pling
design fo r epibenthic surveys a n d im plem ented inverte­
b rate biodiversity m on ito rin g on n a tio n a l groundfish
surveys (G F S ), w hich m o n ito r fish stocks fo r scientific
an d m anagem ent purposes (Jennings et al., 1999; A non.,
2001; C allaw ay et al., 2002). T his greatly extended the
geographical coverage w ith lim ited financial resources.
T he first intern atio n al survey w as ca rrie d o u t in 1999
(Z iihlke e t al., 2001). H ere we re p o rt th e results o f the
2000 survey. A 2 m steel beam traw l (Jennings et a l,
1999) w as deployed d uring five E u ro p e an 3rd q u arter
groundfish surveys to sam ple epibenthos. D iversity of
dem ersal fish com m unities w as assessed from th e catches
o f the 2-m beam traw l as w ell as th e m ain survey traw l a G ra n d e O uverture V erticale (G O V ) o tte r traw l
(A non., 1996). T he aim s o f th e p rese n t study are to
provide a large scale description o f epibenth o s an d fish
diversity th ro u g h o u t the N o rth Sea, to identify spatial
p attern s in fish an d benthic co m m unity stru ctu re an d to
relate environm ental factors, including the frequency
o f traw ling disturbance, to diversity an d com m unity
structure.
M ethods
E p ibenthic invertebrates and dem ersal fish were sam pled
a t 270 statio n s in the N o rth Sea, covering 139 ICES
rectangles (F igure 1). Sam ples were tak en during
n atio n al 3rd q u a rte r 2000 groundfish surveys (G F S ) of
five E u ro p ean countries: G erm any, T h e N etherlands,
E n g lan d , N o rw ay an d D enm ark.
S a m p lin g g e a r a n d p ro c ed u re
Sam ples w ere tak en with a 2-m b eam traw l constructed
fro m galvanized steel. It w as fitted w ith a 20 mm
Epibenthic invertebrates and fish in the N orth Sea
stretched m esh (10 m m knot to knot) an d a co d -en d liner
o f 4 m m k notless m esh (2 m m “ k n o t” to “ k n o t” ). A
chain -m at w as a ttac h ed to minimize the catch o f heavy
rocks. D etails a b o u t th e equipm ent and sam pling p ro to ­
col are given in Jennings et al. (1999) a n d Z ü h lk e et al.
(2 0 0 1 ).
S am p le tr e a tm e n t
Sam ples w ere w ashed through a 5 m m sieve (internal
m esh size) an d epibenthic fauna and fish w ere sep arated
from other m aterial. Species th a t could n o t be identified
a t sea w ere preserved in 4% form alin solu tio n buffered
w ith 3 g I - 1 so d iu m acetate trihydrate for later identifi­
c atio n in th e la b o ra to ry . All anim als were identified to
th e low est possible taxonom ic level. Species nam es w ere
sta n d ard ized to th e nom enclature o f P icto n a n d H ow son
(1999). F ree-living fauna and fish were co u n ted an d
w eighed w ith a seagoing m arine scale (Pols) w ith an
accuracy o f 1 g. Sessile anim als were recorded as p resen t
o r absent.
IC E S fish d a ta
In ad d itio n to th e d a ta collected from th e 2-m b eam
traw l sam ples, species-abundance d ata fo r fish cau g h t
w ith the G O V survey o tte r traw l on the n a tio n a l G F S s
(K nijn e t al., 1993) were analysed. T he d a ta w ere
extracted fro m th e IC E S (International C ouncil fo r th e
E x p lo ratio n o f th e Sea) International B ottom T raw l
Survey D a ta b a se (C openhagen, D K ).
E n v iro n m e n ta l d a ta
Sedim ent sam ples w ere taken a t 60 statio n s d urin g the
English survey. T he dried sedim ent sam ples were sieved
th ro u g h a series o f sta n d ard sieves from 2000 p m to
63 p m m esh to determ ine the grain size d istrib u tio n .
B ottom and surface tem perature and salinity d a ta were
recorded on th e survey vessel an d p rovided fro m a
d atab a se m a in ta in e d by C E FA S (Low estoft, U K ).
T h e latest a n d m o st com plete in tern atio n al co m m er­
cial traw ling effort d a ta were available for 1998.
Six countries c o n trib u te d to a com pilation: T h e
N etherlands, G erm an y , E ngland, S cotland, N o rw ay an d
D enm ark. B eam traw lin g an d o tte r traw ling effort w ere
separated. All d a ta were recorded as hours fishing by
ICES statistical rectangle (boxes 0.5° la titu d e x I o lo n g i­
tude), since we did n o t have sufficiently disaggregated
d ata to a c co u n t fo r differences in vessel type, pow er,
tow ing speed, fishing gear design a n d o th er factors th a t
w ould influence th e are a im pacted by traw ls per u n it
time.
1201
were calculated, w here N 0 is the total n um ber o f species
(species richness) an d N-, is an index o f the n u m b e r o f
a b u n d a n t species [exp (H ), w here H is S h an n o n -W ien er
diversity] (H ill, 1973).
A b u n d an ce a n d b io m ass were standardized to a tow
length o f 200m (a re a = 4 0 0 m 2). N um bers o f species were
analysed as n um bers o f species per haul.
M u ltiv ariate co m m u n ity analysis was carried o u t w ith
the statistical package P R IM E R 5 (C larke an d W arw ick
1994). H ierarchical clu ste r analysis w as used to separate
g roups o f statio n s w ith sim ilar species assem blages. The
d a ta m atrix fo r all ep ib en th ic species w as presence/
absence tran sfo rm ed , b ecause sessile species were
recorded as present o r a b sen t only. A b u n d an ce o f fish
w as d o u b le sq u are ro o t transform ed. T he B ray -C u rtis
index w as calculated betw een each possible p a ir o f
sam ples. Species w hich w ere predom inantly responsible
fo r th e sim ilarity w ithin clusters and for th e sep aratio n
betw een clusters, were determ ined w ith the P R IM E R
p ro g ram S IM P E R . T his exam ines the percentage co n tri­
b u tio n th a t each species m akes to the sim ilarity w ithin
th e cluster a n d to the difference between tw o clusters.
T h e term “ co m m u n ity ” is used for groups o f stations
w ith sim ilar epibenthic o r fish species derived from
cluster analysis. E cological interactions are n o t implied.
T h e relationships betw een environm ental facto rs and
u n iv ariate diversity indices w ere analysed by calculating
p ro d u c t m o m e n t co rrelatio n coefficients. M ultivariate
statistical to o ls w ere then u sed to analyse the epibenthic
co m m unity stru ctu re in rela tio n to environm ental fac­
tors. T h e P R IM E R p ro g ra m BIO -EN V (C lark e and
W arw ick, 1994) calculated w hich set o f environm ental
factors an d fishing effort w as best correlated w ith the
epibenthic com m unity stru ctu re. F o r each possible com ­
b in atio n o f enviro n m en tal factors a dissim ilarity m atrix
based on n o rm alized E uclidean distances w as calculated.
T he agreem ent betw een th e biotic m atrix an d m atrices
o f en v iro n m en tal facto rs w as expressed as the S pearm an
ra n k co rrelatio n coefficient.
T o visualize th e relatio n sh ip between environm ental
factors a n d th e epibenthic com m unity stru ctu re, values
o f abiotic facto rs w ere superim posed onto M D S plots
(m ulti dim ensional scaling). M D S plots were based on
B ray -C u rtis indices o f presence/absence transform ed
epibenthic d ata.
Results
A to tal o f 456 epibenthic a n d 64 fish species was
recorded a t th e 270 stations sam pled during the surveys.
D a ta analysis
D iv e r s ity p a t t e r n o f e p ib e n th ic species
M ean diversity indices an d biom ass were calculated fo r
each ICES rectangle. H ill’s diversity indices N 0 a n d N ,
N u m b ers o f invertebrate species were low er in the
so u th ern th a n in the n o rth e rn N o rth Sea (F ig u re 2a).
R. Callaway et al.
1202
«
M
V
B
2 0 -5 0 0
5 0 0-1000
19-31
1000-8440
3 1 -5 8
(a) E pibenthic species
(b) Biomass epibenthic species
S
12-16
1 6-24
(c) D em ersal fish (2 m beam traw l)
(d) Fish (otter trawl)
F igure 2. T o ta l n u m b e r o f epib en th ic species a n d fish (IC E S rectangle - ') in the N o rth Sea in 2000. (a) T o ta l nu m b er o f epibenthic
species, (b) M e a n epib en th ic biom ass. N um bers are based on w et-w eight o f free living species, (c) T o ta l n u m b e r o f dem ersal fish
caught with a 2-m beam traw l, (d) T o ta l num ber o f fish species (h traw lin g - ‘) caught by otter traw ling. Source o f data: ICES
International B o tto m T raw l S urvey D atabase.
Epibenthic invertebrates and fis h in the N o rth Sea
T he division between areas o f low a n d high num bers o f
species roughly corresponded w ith th e 50 m contour. In
the southern N o rth Sea th e m ean n u m b e r o f species was
14 ± 6 haul - \ while in th e n o rth e rn N o rth Sea it was
30 ± 10 hau l - ’. A n a re a off th e H u m b e r E stuary and
N orfolk coast did n o t fit th e general pattern. Here,
num bers o f species were com p arab le to stations in the
n o rth ern N o rth Sea, alth o u g h th e a re a w as less than
50 m deep.
Biomass was high along th e co n tin en tal coast, in
central p arts o f the N o rth Sea an d a t several stations in
th e n o rth ern N o rth Sea (F igure 2b). A lim ited num ber
o f free-living species w as responsible for th is pattern . In
shallow p a rts o f the N o rth Sea a lo n g the continental
coast, th e starfish A sterias rubens an d the brittle
stars Ophiura albida an d Ophiura ophiura were ab u n ­
d ant. T he herm it crab Pagurus bernhardus and the
starfish A stropecten irregularis w ere responsible for high
biom ass a t stations in th e central N o rth Sea, while in the
n o rth ern N o rth Sea Echinus spp. a n d herm it crabs
con trib u ted to the high biom ass recorded. This is
p artly reflected in the distribution p a tte rn s o f the m ore
a b u n d a n t species (F igure 3).
D iv ersity p a tte rn o f fish species
M o st o f th e fish caught w ith th e 2-m beam traw l were
sm all dem ersal (b o tto m dw elling) species. T he spatial
p a tte rn o f species richness oppo sed th a t o f the epi­
benthos (Figure 2c), w ith th e highest n u m b e rs o f species
in th e southern N o rth Sea an d low num b ers in the n o rth
(w ith the exception o f areas a ro u n d th e Scottish coast).
T he num bers o f fish ca u g h t by o tte r traw ling (source:
IC E S database) were high in the far n o rth , a t some
stations along the co n tin en tal an d E nglish coast an d a t a
few stations in the central N o rth Sea (F ig u re 2d).
E p ib e n th ic c o m m u n ity s tru c tu re
C luster analysis based on presence/absence transform ed
d a ta divided the fau n a in to tw o m a in clusters, which
co u ld be fu rth er subdivided into seven clusters (Figure
4a, T able 1). T he 50 m d ep th c o n to u r divided the two
m ain clusters (Figure 4b). F o u r sub-clusters w ere situ­
ated in the south ern N o rth Sea, a t d ep th s less th a n 50 m,
a n d three in the central an d n o rth e rn N o rth Sea. T he
largest com m unity in th e so u th e rn N o rth Sea (Figure 4,
■ ) was characterized by P. bernhardus (F igure 3a), A.
rubens (Figure 3b), A. irregularis, C orystes cassivelaunus
(F igure 3c), Liocarcinus holsatus, O. ophiura, O. albida
(F ig u re 3d) a n d Psam m echinus m iliaris. T h e m ajority o f
IC E S rectangles along th e co n tin en tal co a st form ed a
g ro u p (Figure 4, • ) , w hich w as a species-poor version
o f th e afore m entioned com m unity. T h e scarcity o f C.
cassivelaunus, P. m iliaris an d A phrodita aculeata, as well
as the absence o f H ydrallm ania fa lc a ta , contribu ted to
1203
th eir sep aratio n . T h e cluster in the so u th ern -m o st area
o f th e N o rth Sea to w ard s the English C hannel (Figure
4, A ) w as characterized by the reg u lar presence o f
th e shrim ps Crangon crangon, Crangon allmanni and
Philoceras trispinosus, w hile C. cassivelaunus a n d A.
irregularis w ere less ab u n d a n t. A lth o u g h sessile species
played a m in o r role in th e so u th ern N o rth Sea, one
region shallow er th a n 50 m d ep th , stretching from
N o rfo lk u p to the D ogger B ank, w as an exception
(F ig u re 4, ♦ ) . T his cluster w as sep arated from the
o th ers by th e reg u lar presence o f bryozoans such as
A lcyonidium diaphanum (F ig u re 3j), A lcyonidium para­
siticum , E ucratea loricata a n d Flustra foliacea as well as
h y d ro zo an s such as H. fa lc a ta (Figure 3i) an d Halecium
halecium. In this respect th is cluster w as sim ilar to the
n o rth e rn N o rth Sea cluster, alth o u g h the free-living
fa u n a were typical so u th ern N o rth Sea species.
C lusters in th e n o rth ern N o rth Sea were located
betw een the dep th co ntours 50-100 m an d 100-200 m,
b u t th e ir sep aratio n w as less conspicuous th a n between
clusters in th e so u th (Figure 4). A large range o f species
fo u n d betw een th e 50 m a n d 100 m d ep th (F igure 4, O ),
w ere n o t reco rd ed in the shallow er so u th ern N o rth Sea.
Species co n trib u tin g p red o m in an tly to the sep aratio n o f
this cluster fro m the ones to the so u th w ere w helks such
as N eptunea antiqua (Figure 3f) a n d Colus gracilis, the
h erm it crab s Pagurus pubescens an d Anapagurus laevis
(F igure 3g) as well as species such as H ydroides nor­
vegica, H ya s coarctatus (F igure 3e), F. foliacea and
E pizoanthus papillosus (form erly E. incrustatus). The
cluster betw een th e 100 m a n d 200 m d ep th line (Figure
4, O ) w as characterized by A. irregularis, Hyalinoecia
tubicula (F igure 3h), H orm athia digitata (Figure 3k) and
E chinus spp. M an y species occurring regularly in
th e 50-100 m com m unity w ere less frequently found
betw een 100 m an d 200 m (e.g. A. digitatum , F. foliacea
a n d Tubularia indivisa), a facto r co n trib u tin g to the
se p aratio n o f th e clusters.
A sm all cluster w as form ed by the IC E S rectangles
n o rth o f 61°N la titu d e (Figure 4, □ ) . T hese stations
w ere ch aracterized by the presence o f th e hexacoral
Caryophyllia sm ithii (Figure 31) b u t otherw ise by the
absence o f several species fo u n d fu rth er south.
F ish c o m m u n ity stru c tu re
F ish com m unities w ere divided in to tw o m ain groups
located so u th an d n o rth o f the 50 m d ep th line. T his was
tru e fo r fish ca u g h t w ith the 2-m b eam traw l and otter
traw l. C luster analysis based on fish species caught with
th e 2-m b eam traw l separated one cluster for the so u th ­
ern N o rth Sea an d three for th e n o rth (F igure 5a & b,
T ab le 2).
T h e co m m u n ity o f the so u th e rn N o rth Sea (Figure
5b, ■ ) w as characterized by sm all, non-com m ercial
species. M o st com m on w ere Buglossidium luteum
R. Callaway et al.
1204
----------------------- i n a n t , »A
o
—- »X © 151-1
-1—
507
• 107-4i7
XT3
/ 'iJ Ê f
1-10
10-50
# 50-4240
9
O
,* ? ] •
1-5
5-15
15-22
i r X
j
W\
-X
X
O
«■•?!
x
.
j f
C
% fJ Ë à
"F
y
--- -
Æ
e 1-10
X O10-70
V» # 70-80
n r
J
*
(b) A s te r ia s ru b en s
(c) C orystes ca ssivela u n u s
^ g y L |p J
x¡&-“A%6% ©0 1no
0-40
X J k j X i » 40-100
I S
S
S
Vl
uÎXrÂ
* °
X
}
r g Æ
I
f
tm
m
e0.1-14
14-50
• so-ios
i r a
r
t#}l *
(h) H yalinoecia tubicola
i§ £
,—-jA.
%
o
^ °
%
p
Æ>
(k) H o r m a th ia d ig ita ta
g®0.1-2
2-10
(1) C aryophyllia s m ith ii
X
^ \ [
®° 4-32
.-«I?] #
32-54
,
}$ *30
p
(j) A lc y o n id iu m d ia p h a n u m
©
a
iv ? ) ( * * *** / ^ c
ss¿4~
/— X
.X .
X
P
- ^
(g) A n a p a g u ru s laevis
V
(i) H y d r a llm a n ia falcata
^/TT^s © 20-500
# 500-2350
M. P
-4 j £
\
C s íí*
000
(d) O phiura a lbida
. ¿
(f) N e p tu n e a a n tiq u a
^ X . O 20-500
# 500^ 8
^
0
J
yf
(e) H y a s coarctatus
s ffa P
/ • » .# » #“••Ir«SX<t
*
i
(a) P a g u ru s bernhardus
V xi
'-
s
X
C
r
(
s*
c7 * r
a éÊ yh
*
(m) B u g lo s sid iu m Luteum
(n) L im a n d a lim a n d a
(o) H ippoglossoides
pla tesso id es
(p) M yxine g lu tin o sa
F ig u re 3. D istrib u tio n o f species c h ara cte ristic for epibenthic and fish com m unities. T he size o f bubbles indicates abundance.
Species w ere reco rd ed as present o r a b se n t if no legend is show n.
Epibenthic invertebrates and fis h in the N o rth Sea
1205
E pibenthic com m unities
▲
♦
o
o
LJ
B ray -C u rtis sim ilarity
i%)
F ig u re 4. E pib en th ic com m unities in th e N o rth Sea in 2000. (a) H ierarchical classification analysis based o n presence/absence
tran sfo rm e d d a ta . V ertical bars indicate th e m ain groupings, (b) L o ca tio n o f clusters identified by h ierarchical classification
analysis (☆ outliers).
(solenette, F igure 3m), Lim anda limanda (dab, F igure
3n) a n d Callionymus lyra (dragonet). Arnoglossus laterna
(scaldfish) w ere regularly recorded, b u t in low num bers.
In th e ce n tral a n d n o rth ern N o rth Sea one cluster
ranged from 50-100 m and an o th e r from 100-200 m.
Betw een 50 m and 100 m (F igure 5b, A ) few er o f the
sm all, non-com m ercial species were fo u n d . N um bers
o f L. lim anda w ere higher th a n in th e so u th an d
H ippoglossoides platessoides (long rough dab, F igure 3o)
were recorded regularly. In w aters deeper th a n 100 m
(F ig u re 5b, O ), H. platessoides w as th e characterizing
species fo r th e com m unity, to g eth er w ith M yxin e gluti­
nosa (hagfish, F ig u re 3p). Som e IC E S rectangles in the
far n o rth (F igure 5b, O ), a ro u n d th e 200 m dep th line,
were g ro u p ed as a th ird cluster based on the pau city o f
dem ersal fish in this area.
F o r fish ca u g h t w ith the G O V traw l d uring the survey
(source: IC E S d atabase) cluster analysis form ed three
clusters in th e so u th , one in th e cen tral a n d one in the
n o rth e rn N o rth Sea (Figure 5c & d, T able 3).
In th e so u th ern N o rth Sea two o f th e three clusters
w ere spatially dom inating, b u t th ey were in te r­
m ixed ra th e r th a n geographically discrete (Figure 5d,
■ © ). B oth clusters were characterized by a sim ilar
ran g e o f species, including M erlangius m erlangus (whit­
ing), E utrigla gurnardus (grey g u rn ard ), L. limanda an d
Trachurus trachurus (scad). Differences in ab u n d an ce o f
these species w ere responsible for th e sep aratio n s into
tw o clusters. A th ird smaller cluster w as form ed in
the fa r so u th betw een the English an d D u tch coast
(F ig u re 5d, A ). H igher num bers o f T. trachurus an d
Scom ber scom brus (mackerel) were recorded in this
area.
A s w ith the analysis based on th e 2-m beam -traw l
sam ples, th ere w as a division betw een com m unities from
50-100 m d ep th (Figure 5d, O ) an d 100-200m d ep th
(F ig u re 5d, □ ) . T h e area between th e 50 m an d 100 m
c o n to u r w as num erically dom inated by M elanogram m us
aeglefinus (haddock), M. merlangus, Clupea harengus
(herring) an d Pleuronectes platessa (plaice). Between
R. Callaw ay et al.
1206
T able 1. E p ib e n th ic species w hich a cc o u n t for m o st o f the sim ilarity w ithin th e clusters identified by
cluster analysis. T he sym bols fo r clusters follow F ig u re 4. “ M ean sim ilarity” is the m ean sim ilarity
w ithin th e clu ster a n d “ c o n trib u tio n ” is th e c o n trib u tio n o f individual species to the to ta l sim ilarity
w ithin the g roup. Species c o n trib u tin g a lto g e th e r 60% o f the sim ilarity o r th e six species contributing
m o st to th e sim ilarity are shown.
C lu ster A
M ean sim ilarity: 44.84
Pagurus bernhardus
L iocarcinus holsatus
Crangon allm anni
Ophiura albida
A sterias rubens
Crangon crangon
C um ulative
c o n trib u tio n
to sim ilarity
(%)
10.30
20.59
30.89
39.63
47.79
54.39
C lu ster <>
M e a n sim ilarity: 42.10
O phiura ophiura
P agurus bernhardus
Liocarcinus holsatus
A lcyonidium diaphanum
E lectra pilosa
H ydrallm ania fa lc a ta
P agurus bernhardus
Liocarcinus holsatus
A sterias rubens
Ophiura ophiura
23.08
46.16
69.24
77.52
C lu ster 11
M e a n sim ilarity: 45.08
6.29
12.57
18.86
25.14
30.35
35.50
C lu ster O
M e a n sim ilarity: 42.12
P agurus bernhardus
N eptunea antiqua
A sterias rubens
C olus gracilis
H y a s coarctatus
H ydractinia echinata
C lu ste r ®
M e a n sim ilarity: 49.90
C um ulative
contrib u tio n
to sim ilarity
(%)
P agurus bernhardus
A sterias rubens
A stropecten irregularis
C orystes cassivelaunus
Liocarcinus holsatus
O phiura ophiura
9.29
18.57
27.86
36.69
43.63
49.43
C lu ster O
M e a n sim ilarity: 39.85
4.73
9.19
13.33
17.28
20.96
24.59
A stropecten irregularis
H yalinoecia tubicola
H orm athia digitata
E chinus sp.
Anapagurus laevis
P agurus pubescens
5.37
10.22
14.89
19.44
23.93
27.84
C lu ster □
M e a n sim ilarity: 41.15
E pizoanthus papillosus
E chinus sp.
Anapagurus laevis
A dam sia carciniopados
C aryophyllia sm ithii
P andalus m ontagui
10.14
20.28
30.42
40.56
46.07
51.50
100 m a n d 200 m Trisopterus esm arkii (N orw ay p o u t)
was the d o m in a n t species in th e com m unity.
E n v iro n m e n ta l fa c to rs a n d fishing effort
E nvironm ental factors w ere significantly correlated,
both positively a n d negatively, w ith univariate diversity
indices o f th e epibenthic fau n a or fish (T able 4).
N um bers o f epibenthic species were negatively co rre­
lated w ith b o tto m te m p e ra tu re an d beam traw ling effort
(Table 4, F igure 6), w hile they were positively correlated
with b o tto m salinity.
Few er correlations w ith environm ental factors were
fo u n d when analysing the so u th ern N o rth Sea (< 50 m
d ep th ) an d cen tral-n o rth ern N o rth Sea (>50 m depth)
separately, possibly due to low er num bers o f samples
reducing the pow er to detect statistical differences. In
th e sou th ern N o rth Sea num bers o f sessile benthic
species were negatively co rrelated w ith beam traw ling
effort, b u t there were no o th er significant correlations
fo r the benthos. In the n o rth e rn N o rth Sea num bers
o f epibenthic species w ere correlated w ith several
sedim ent param eters a n d b eam traw ling effort
(T able 1).
F is h c a u g h t w ith 2 m b eam traw l
1' o
0
10 20 30 40 50 60 70 80 90 100
B ray-C urtis sim ilarity (%)
Fish caught w ith otter traw l
A
■
O
B ray-C urtis sim ilarity (Of )
Figure 5. F ish com m unities in th e N o rth S ea in 2000. (a) H ierarchical classification analysis based on 4th ro o t transform ed d ata
o f fish caught w ith a 2-m beam traw l. V ertical bars in d icate m ain groupings, (b) L o ca tio n o f fish com m unities indicated by
d endrogram (a), (c) H ierarchical classfication analysis based on 4 th ro o t transform ed d a ta o f fish c au g h t by o tte r traw ling during
groundfish surveys (source o f data: IC E S d atabase), (d) L o c a tio n o f fish com m unities in d ic a te d by den d ro g ram (c).
R. Callaway et al.
1208
T able 2. F ish species caught w ith th e 2-m b e am traw l, w hich accounted fo r m o st o f th e sim ilarity
w ithin th e clusters identified by clu ster analysis. T h e sym bols for clusters follow F ig u re 5. “ M ean
sim ilarity” is the m ean sim ilarity w ithin th e cluster a n d “ c o n trib u tio n ” is th e c o n trib u tio n o f individual
species to th e to tal sim ilarity w ithin th e group. Species contributing alto g eth er 70% o f th e sim ilarity are
show n.
C luster □
M e a n sim ilarity: 46.36
C um ulative
c o n trib u tio n
to sim ilarity
(%)
Buglossidium luteum
L im anda limanda
C allionym us lyra
P om atoschistus minutus
21 Á l
47.09
59.48
70.28
C luster A
M e a n sim ilarity: 37.88
L im anda limanda
H ippoglossoides platessoides
C luster O
M ean sim ilarity: 36.05
Hippoglossoides platessoides
M y xin e glutinosa
C um ulative
contribution
to sim ilarity
(%)
58.54
71.86
C luster O
M ean sim ilarity: 35.93
48.16
72.18
Gadiculus argenteus
100.00
T ab le 3. F ish species, caught by o tte r traw ling (m ain groundfish survey), w hich acco u n te d for m ost
o f th e sim ilarity w ithin th e clusters identified by c luster analysis. T he sym bols fo r clusters follow Figure
5. “ M e a n sim ilarity” is th e m ean sim ilarity w ithin th e cluster and “ c o n trib u tio n ” is th e c o n trib u tio n o f
individual species to th e to ta l sim ilarity w ithin th e g roup. Species c o n trib u tin g to a lto g e th e r 70% o f the
sim ilarity are shown.
C lu ster A
M ean sim ilarity: 47.72
C um ulative
c o n trib u tio n
to sim ilarity
(%)
C luster S
M e a n sim ilarity: 59.55
C um ulative
contrib u tio n
to sim ilarity
(%)
M erlangius merlangus
Trachurus trachurus
L im anda limanda
Scom ber scombrus
Pleuronectes platessa
E chiichthys vipera
19.58
32.73
43.95
51.68
59.12
65.75
Sprattus sprattus
M erlangius merlangus
Lim anda limanda
Clupea harengus
Trachurus trachurus
E utrigla gurnardus
19.94
37.09
48.11
58.21
64.15
69.57
C luster ®
M ean sim ilarity: 58.31
L im anda limanda
M erlangius merlangus
E utrigla gurnardus
Trachurus trachurus
Scom ber scombrus
Pleuronectes platessa
C luster O
M e a n sim ilarity: 69.51
19.66
34.21
47.40
56.48
64.46
72.31
C lu ster □
M ean sim ilarity: 65.57
Trisopterus esm arkii
M elanogram m us aeglefinus
M erlangius merlangus
Clupea harengus
Pleuronectes platessa
M icrostom us k itt
22.63
39.64
50.52
59.67
67.72
71.99
M elanogram m us aeglefinus
M erlangius merlangus
Lim anda limanda
Pleuronectes platessa
Clupea harengus
E utrigla gurnardus
20.39
34.66
A IM
56.56
65.05
72.99
Epibenthic invertebrates and. fis h in the North Sea
(a) Beam traw ling (h y e a r-1)
1209
(b) O tter traw ling (h y ea r-1)
«
{Ü
1-5000
5000-25 000
5 0 0 0 -2 5 000
25 0 0 0-5 8 200
25 0 0 0 -4 5 300
F ig u re 6. B eam and o tte r traw lin g effort 1998 (h ICES rectangle
E ngland, D enm ark, N o rw ay a n d S cotland.
T h e BIO -EN V p ro ce d u re (C lark an d W arw ick, 1994)
w as used to find the c o m b in atio n o f abiotic factors
w hich best accounted fo r th e epibenthic com m unity
stru ctu re (Table 5, F ig u re 7). In th e entire N o rth Sea
a n d in the n o rth ern N o rth Sea, b o tto m tem perature, the
m u d content o f the sed im en t a n d beam traw ling p r o ­
vided the best explan atio n o f epibenthic com m unity
stru ctu re (Figure 7, T ab le 5). In th e southern N o rth Sea
th e p a tte rn s in com m unity stru ctu re w ere best explained
by o th er sedim ent p ara m ete rs, beam traw ling effort an d
o tte r traw ling effort (T a b le 5).
T his analysis w as com prom ized by several in te r­
correlations betw een env iro n m en tal factors. N egative
correlations were fo u n d fo r beam traw ling an d o tte r
traw ling, b o tto m te m p e ra tu re an d th e am ount o f m u d
in th e sedim ent, the 250 p m a n d 125 pm sedim ent frac­
tio n fo r stations in less th a n 50 m d e p th and b o tto m
te m p eratu re and b o tto m salinity fo r statio n s deeper th a n
50 m . Positive correlations occurred betw een b o tto m
tem p eratu re and beam traw lin g a n d beam traw ling an d
the a m o u n t o f fine sa n d in the so u th e rn N o rth Sea.
C larke an d W arw ick (1994) suggested substituting
correlated factors by ju s t o ne o f them . H ow ever, we
1 year
'). C um ulated d a ta o f G erm any, T he N etherlands,
decided to include all factors in the analysis because the
in ter-co rrelatio n s between environm ental facto rs were
n o t co n sisten t fo r all the sp atial areas analysed.
Discussion
T h e fa u n a o f th e southern N o rth Sea w as d om inated by
free-living invertebrates and sm all dem ersal fish, while in
th e n o rth sessile species played a greater role. A hybrid
area, w ith a sessile fauna typical fo r th e n o rth e rn N o rth
Sea an d free-living species ch aracteristic fo r th e south,
w as fo u n d o ff th e coast o f N o rfo lk an d F lam b o ro u g h
stretch in g to w ard s the D ogger Bank.
T h e m o st conspicuous b o u n d ary fo r N o rth Sea fish
a n d ep ib en th ic com m unities w as the 50 m d ep th line
w hich sep arates the southern a n d cen tral-n o rth ern
N o rth Sea. T h e 100 m and 200 m co n to u rs can also be
view ed as fau n al boundaries, b u t they are less clearly
defined th a n th e boundary at 50 m depth. T hese b o u n d ­
aries are sim ilar to those noted by K iinitzer et al. (1992)
in th eir stu d y o f infauna com m unities.
R. Callaway et al.
1210
T able 4. Significant positive a n d negative correlations (p ro d u ct m o m en t correlation coefficient (r),
p < 0.05) betw een u n iv ariate indices o f the fauna a n d en v iro n m en tal fa cto rs (sedim ent fraction
1000 pm , 500 pm , 250 pm , 63 pm , < 63 pm , b o tto m tem p eratu re, b o tto m salinity, otter traw ling, beam
traw ling). C o rre la tio n analysis w as carried o u t fo r statio n s o f th e entire N o rth Sea (A), fo r stations
shallow er than 50 m (B) and deeper than 50 m (C). I f a n environm ental factor is n o t listed no
significant relatio n sh ip w as found. N u m b e rs o f fish species w ere derived fro m 2-m beam traw l sam ples.
T he analysis w as based on sam ples o f the English survey.
A. all sam pling
stations; n= 59,
r0.05(2),57= 0.256
Sedim ent <63 pm (g)
B ottom tem p eratu re (°C)
B o tto m salinity (%o)
Beam traw ling ( h y _ 1 )
T o ta l
n u m b er o f
epibenthic
species
N u m b er o f
sessile
epibenthic
species
N um ber of
free-living
epibenthic
species
H ill’s N ,
(based on
free-living
epibenthos)
- 0 .6 7
+ 0.50
- 0 .5 5
- 0 .3 1
- 0 .5 0
+ 0.37
- 0 .5 0
- 0 .6 7
+ 0.50
- 0 .4 5
- 0 .5 4
N um ber o f
fish
species
+0.43
- 0 .2 6
- 0 .4 1
B. stations <50 m
depth; n = 2 1 ,
r0.05(2), 19=0.433
Sedim ent 1000 pm (g)
B o tto m tem p eratu re (°C)
B eam traw ling (h y ~ *)
+ 0.56
+ 0.44
- 0 .4 4
C. statio n s >50 m
dep th ; n = 3 8 ,
r0.05(2),36= 0.320
Sedim ent <63 pm (g)
Sedim ent 63 pm (g)
Sedim ent 125 pm (g)
Sedim ent 250 pm (g)
B o tto m tem perature (°C)
B eam traw ling (h y ~ l )
- 0 .4 7
+0.38
- 0 .6 4
- 0 .6 0
+ 0.43
+ 0.51
- 0 .4 7
- 0 .3 2
E p ib e n th o s
T he low er epibenthic diversity a n d species richness in the
south ern th a n n o rth e rn N o rth Sea w as consistent with
th e results o f previous studies (F rau e n h eim e t al., 1989;
Jennings et al., 1999; Z iihlke et al., 2001). However,
previous au th o rs did n o t se p arate th e ep ibenth o s o f the
so u th into different com m unities, m ainly d u e to the
constraints im posed by low levels o f sam ple replication
a n d the statistical tools availab le a t th a t tim e. We
differentiated fo u r com m unities in the so u th e rn N o rth
Sea. T heir se p aratio n w as based o n the rich sessile fauna
o ff the N o rfo lk and F la m b o ro u g h coasts, high
n um bers o f shrim p and praw n species in th e fa r south
a n d the overall scarcity o f species alo n g th e co n tinental
coast.
Several species w ere u b iq u ito u s in all fo u r com m uni­
ties, such as th e brittle star Ophiura ophiura, th e starfish
Asterias rubens, th e herm it c ra b Pagurus bernhardus and
th e swimm ing crab Liocarcinus holsatus. T h eir ubiquity
w as consistent w ith p atterns described in prev io u s years
(e.g. F rauenheim et a l, 1989; Jen n in g s et a l , 1999; Rees
et a l, 1999).
- 0 .3 3
W e identified three epibenthic com m unities in the
ce n tral-n o rth ern N o rth Sea, an d again, the locations o f
these com m unities are b roadly co n sisten t w ith those
described in previous reports (D yer e t a l , 1983; B asford
et a l, 1989; F rau en h eim et a l, 1989; Jennings et a l,
1999). T h e 50 m , 100 m an d 200 m boundaries aiso
coincide w ith th e concept o f th ree infaunal étages
developed by G lem arec (1973), as confirm ed and
differentiated by K ünitzer et al. (1992).
T h e 50 m d ep th co n to u r was th e so u th ern distribution
b o rd er fo r m an y sessile an d free-living species. H ow ever,
th e ab u n d a n ce o f free-living fa u n a in the 50-100 m
co m m u n ity w as d om inated by species which w ere also
fo u n d in th e so u th , nam ely P. bernhardus, A. rubens an d
A. irregularis. T his w as in accordance w ith the results of
D y er et a l (1983), w ho stated th a t m a n y species which
were co m m o n a t th e sou th ern sta tio n s were cau g h t in
th e n o rth . M o st conspicuous in th e com m unity fo u n d at
depths o f 50-100 m was th e diverse sessile fau n a, w ith
n u m ero u s hydrozoans, bryozoans a n d tube-dw elling
polychaetes. B asford et al. (1989) fo u n d m atching d istri­
b u tio n p a tte rn s fo r several o f the m o st com m on sessile
species (e.g. Alcyonidium digitatum , Flustra foliacea,
1211
Epibenthic invertebrates and fis h in the North Sea
T ab le 5. R elationship betw een en v iro n m en tal factors a n d th e c om m unity structure. T he c o m b in a tio n o f e nvironm ental factors,
w hich show ed th e highest correlatio n [Spearm an ra n k c o rre latio n (p)] w ith th e epibenthic c o m m u n ity stru c tu re are listed. Results
a re based on m ultiv ariate d a ta analysis.
E n tire N o rth Sea
< 50 m d ep th (south)
> 50 m d e p th (north)
A ll epibenthic species
© B o tto m tem p eratu re
o < 63 p m sedim ent fraction
© B eam traw ling
p= 0.67
• 250 pm sedim ent fraction
o 125 pm sedim ent fraction
• Beam traw ling
o O tte r traw ling
p = 0 .5 0
® B ottom tem p eratu re
o B ottom salinity
• 125 pm sed im en t fraction
© <63 pm sedim ent fraction
• B eam traw ling
p = 0 .5 8
Free-living species
a B o tto m tem perature
• < 63 p m sedim ent fraction
e Beam traw ling
p=0.61
« 1000 p m sedim ent fraction
• 500 p m sedim ent fraction
® 250 pm sedim ent fraction
• B o tto m salinity
a O tte r traw ling
p = 0 .5 6
o 500 pm sedim ent fraction
• 63 p m sedim ent fraction
© B ottom tem p e ra tu re
o B ottom salinity
© B eam traw ling
p = 0 .5 2
Sessile species
© B o tto m tem perature
© <63 p m sedim ent fraction
© Beam traw ling
p = 0 .6 0
• 250 pm sedim ent fraction
• 125 p m sedim ent fraction
» Beam traw ling
® O tte r traw ling
p = 0 .3 3
• B ottom tem p eratu re
© B ottom salinity
© 63 pm sed im en t fraction
© < 63 pm sedim ent fraction
® B eam traw ling
p= 0 .5 8
Bolocera tuediae) in this area betw een 1980 a n d 1985.
N o rth o f th e 100 m co ntour, E chinus spp. w ere highly
a b u n d a n t a t several stations, as also reported by D yer
et al. (1983).
O u r results w ere also consistent w ith those o f previous
studies w hen viewed a t a sm aller spatial scale. T hus
D uineveld et al. (1991) separated a n epibenthic co m m u ­
n ity situated along th e co n tin en tal coast from an off­
shore com m unity an d other a u th o rs fo u n d a species rich
sessile epifauna off th e N o rfo lk co ast (H am ond, 1969;
R ees et al., 1999). T he presence o f th e sessile bryo zo an
A. diaphanum in the D ogger B ank are a w as m entioned
by D uineveld e t al. (1991) an d D y er e t al. (1983) (both
a u th o rs used th e previous n am e A. gelatinosum ).
F o r th e area n o rth o f th e 100 m co n to u r there are
som e differences betw een ou r results an d those o f p re­
vious authors. F o r exam ple, D y er et al. (1983) an d
B asford e t al. (1989) allocate th e ir n o rth ern sam ples to
r a th e r different clusters. A n ex a m in atio n o f individual
species d istributions suggests th a t these differences
betw een clusters do n o t reflect changes in the spatial
d istrib u tio n o f epibenthic com m unities over tim e.
R a th e r, they are likely to be th e result o f different
sam pling m ethods a n d d ata analysis. S tand ard izatio n o f
epibenthic sam pling techniques w as attem pted only
recently (Jennings e t al., 1999; C allaw ay et a l, 2002).
F ish
A lth o u g h th e tw o sam pling m e th o d s, th e 2-m beam and
o tte r traw l, w ould have selected fo r fish o f different sizes
a n d different species, th e rep o rted sp atial distributions
o f fish com m unities are rem a rk a b ly sim ilar. A s w ith the
epibenthic com m unities, th e spatial d istrib u tio n o f fish
com m unities w as closely rela ted to the < 5 0 m , 50-100 m
a n d 100-200 m depth bands.
H ow ever, because th e tw o types o f traw l w ere catch­
ing different species in th e com m unity, p attern s of
species richness were different. W ith th e 2-m beam trawl,
w hich catches small species closely associated w ith the
seabed, we recorded high species-richness in shallow
are as o f th e sou th ern N o rth Sea an d alo n g th e English
an d S cottish coast an d low species richness in the
n o rth e rn N o rth Sea. W ith th e G O V o tte r traw l, which
m o st effectively catches dem ersal species m o re loosely
associated w ith the seabed (S p arh o lt, 1990), species
richness w as highest in th e far n o rth , in scattered areas
alo n g the English and co n tin en tal coast, an d in the
cen tral N o rth Sea. T his em phasizes th e necessity to
in teg rate results from different survey m eth o d s in order
to gain a com prehensive p icture o f the fish fauna.
T h e fish com m unity p attern s w ere consistent with
th o se described by R ogers et al. (1998), w ho collected
sam ples w ith a 4-m beam traw l. R ees et al. (1999)
in cluded fish cau g h t w ith a 2-m b eam traw l in to their
m u ltiv ariate analysis o f b en th ic co m m u n ity structure
an d show ed th a t small flatfish ch aracterized the south­
ern N o rth Sea benthic assem blage. T h e sm all flatfish
th a t d o m in ate th e southern N o rth Sea b o tto m dwelling
fish co m m unity are significant consum ers o f benthic
in fau n al p ro d u ctio n (B raber & de G ro o t, 1973; Connell
an d A n d erso n 1999).
1212
R. Callaway et al.
32
(C )
• o
.* > • : -o .
<*> b
V
*
o 0*0
•
.QVi ft'»
T em p eratu re
S e d im e n t < 63 pm
(d)
ôo
B e am traw lin g
O tte r tra w lin g
Figure 7. E n v iro n m e n ta l variables linked to th e epibenthic com m unity stru ctu re, (a) M D S o f epibenthic d a ta , E nglish survey data,
presence/absence d a ta tran sfo rm a tio n , stress 0.18. N u m b e rs show th e IC E S line a sam ple w as taken; b -e , environm ental factors
superim posed o n M D S. F actors b -d , best explained th e o rd in atio n (B IO E N V c o rre latio n analysis).
E n v iro n m e n ta l fa c to rs
S edim ent p aram eters, b o tto m te m p eratu re a n d fishing
effort w ere all closely correlated w ith species richness,
diversity a n d com m unity patterns.
S edim ent prop erties are widely regarded as factors
influencing th e benthic fauna (B asford et al., 1990;
K ühne a n d R a ch o r, 1996; Rees et a l , 1999), alth o u g h
the rela tio n sh ip is clearer for in fau n a th a n fo r epifauna
(B asford e t a l , 1990; D uineveld et al., 1991). T h e small
g rab used to sam ple sedim ent m ay provide a n im p res­
sion o f h a b ita t type th a t com prom ises ou r attem p ts to
relate sed im en t properties to the com position o f faunal
sam ples tak en w ith 2-m beam traw ls. T here are two
reaso n s for this. F irst, the grab does n o t effectively
sam ple h a rd su b stra tu m such as gravel, rocks o r dead
shell, w hich is essential as attachm ent m a terial for sessile
species. Second, th e scales at w hich th e beam traw l
sam ples are collected, an d the scales a t w hich epibenthic
species are likely to “ perceive” h ab itat, are orders o f
m a g n itu d e larger th a n the scale a t w hich g rab samples
a re tak en . T o b etter relate h ab itat stru ctu re to epi­
b en th ic a n d fish sam ples taken w ith a 2-m beam trawl
req u ires h a b ita t descriptions th a t in teg rate small-scale
heterogeneity. A coustic m ethods m ay pro v id e a m eans
by w hich to achieve th is goal. B ottom tem p eratu re has a
Epibenthic invertebrates a n d fis h in the N o rth Sea
pow erful influence on th e d istrib u tio n , ab u n d a n ce and
p ro d u ctio n o f benthic fau n a (e.g. D uineveld et a l, 1991;
Brey, 1999). A nnual variation in b o tto m w ater tem pera­
ture is m o st likely to be responsible for th e distinctive
bo u n d ary betw een com m unities found south a n d n o rth
o f 50 m depth . This depth closely m atches th e b o u n d ary
between th e therm ally stratified w aters o f th e northern
N o rth Sea a n d the perm anently vertically m ixed w aters
o f th e so u th e rn N o rth Sea (B row n et a l , 1999). Tidal
currents m a in ta in a turb u len t regim e in th e southern
N o rth Sea, while in the n o rth e rn N o rth Sea th e m ixing is
w eaker a n d th e w ater colum n stratifies in response to
seasonal heatin g (Sim pson, 1994). T his causes higher
seasonal fluctuations in b o tto m w ate r te m p eratu re in the
so u th th a n in th e north.
C urren ts m ay also determ ine boun d aries between
benthic inverteb rate and fish com m unities. L ong-term
circulation o f the N o rth Sea is anticlockw ise (Sim pson,
1994). H ow ever, this is n o t a basin-w ide circulation. In
the n o rth ern N o rth Sea, th e anticlockw ise flow tak es the
fo rm o f a well-defined cu rren t, th e D ooley C urrent,
w hich follow s th e 100 m c o n to u r (T urrell et a l , 1992).
A nother inflow from the A tla n tic enters th e N o rth
Sea from n o rth o f Shetland a n d follow s the 200 m
co n to u r alo n g th e w estern edge o f th e N orw egian
D eeps (D ooley, 1974). T hese cu rren ts m ay influence
pro d u ctio n , larval and post-larv al d istrib u tio n a n d may
acco u n t for som e o f the com m unity p attern s w e observe.
B eam traw lin g is a source o f w idespread an d chronic
disturbance in th e N o rth Sea. E xperim ental studies and
com parisons betw een areas subject to different fre­
quencies o f traw ling disturbance h av e show n th a t beam
traw ling leads to significant m o rta lity o f ben th ic fauna
(Bergm an a n d van Santbrick, 2000), reductions in the
abu n d an ce o f species w ith large b o d y size (R u m o h r and
K ujaw ski, 2000) an d reductions in p ro d u ctio n (Jennings
et a l , 2001). T hese im pacts are reflected in changes in
com m unity stru ctu re (e.g. B ergm an a n d H up, 1992; F rid
an d H aii, 1999; F rid et a l, 2000; L indeboom and
D eG ro o t, 1998). Sessile species are p articu larly vulner­
able to th e im pacts o f traw ling (e.g. C ollie et a l , 1997;
K aiser et a l , 1998, 2000; Freese et a l , 1999). O u r results
suggest th a t beam traw ling effort w as consistently cor­
related w ith ben th ic com m unity stru ctu re , species rich­
ness and diversity. H ow ever, w hile F rid et al. (2000) and
R u m o h r an d K ujaw ski (2000) have show n th a t long­
term increases in th e frequency o f traw ling disturbance
have accounted fo r tem poral changes in the stru ctu re of
N o rth Sea ben th ic com m unities, th e results o f our
sp atial com parisons am ong ben th ic com m unities in
areas subject to different levels o f traw ling distu rb an ce
a re m ore difficult to interpret. T his is because large-scale
p attern s o f b eam traw ling effort in th e N o rth Sea reflect
th e availability o f substrata th a t a re suitable fo r beam
traw ling (e.g. R ijn sd o rp et al., 1998). Such su b stra ta are
likely to be relatively m obile sedim ents th a t do n ot
1213
necessarily p ro v id e g o o d environm ents fo r the develop­
m en t o f a species rich and a b u n d a n t sessile fauna.
M oreover, th e n a tu ra l disturbance th a t results from
w ave an d tid a l actio n on m obile sedim ents will fav o u r
com m unities d o m in a te d by free-living ra th e r th a n sessile
species. D isentangling the relative roles o f fishing, h ab i­
ta t type a n d n a tu ra l disturbance in influencing the
large-scale stru c tu re o f benthic com m unities will require
sp atial an d te m p o ral com parisons am o n g com m unities
in sim ilar physical environm ents th a t are subject to
different levels o f traw ling disturbance. T o date, such
co m p ariso n s h av e involved trade-offs between spatial
scale, te m p o ral scale an d replication w ithin h ab itat
types. N o w th a t we have show n th a t large-scale benthic
surveys ca n be ru n in conjunction w ith th e exist­
ing annual fisheries surveys, the an n u a l collection of
a d d itio n a l b en th ic sam ples within h a b ita t types w ould
allow us to exam ine the effects o f traw ling on
u n p reced en ted sp atial a n d tem poral scales.
Acknowledgements
W e w ish to th a n k th e officers an d crew o f the various
research vessels involved in this survey for their
co -o p eratio n . M a n y thanks to B rian R ack h am an d
R ich a rd A yers fro m C E F A S for providing th e ICES fish
d a ta an d en v iro n m en tal d ata. Two an onym ous referees
m a d e v aluable com m en ts on the first version o f the
m an u scrip t. T h is study w as funded by th e E uropean
C o m m u n ity ( D G IV X , 98/021).
References
A n o n . 2001. R e p o rt o f th e w orking group on th e assessm ent o f
dem ersal sto ck s in th e N o rth Sea and Skagerrak. ICES C M
2001/A C FM : 07.
A n o n . 1996. M a n u a l for the Intern atio n al B ottom T raw l
Surveys. R evision V. A ddendum to IC E S C M 1966/HI.
B asfo rd , D. J., E leftheriou, A ., and Raffaelli, D . 1990. T he
in fa u n a a n d e p ifa u n a o f the n o rth ern N o rth Sea.
N e th erlan d s J o u rn a l o f Sea Research, 25(1/2): 165-173.
B asford, D. J., E leftheriou, A., and Raffaelli, D . 1989. T he
e p ifa u n a o f th e n o rth e rn N o rth Sea (5 6 -6 1°N). Jo u rn a l o f the
M a rin e B iological A ssociation U .K ., 69: 387-407.
B ergm an, M . J. N ., a n d H up, M. 1992. D irect effects o f
beam traw ling on m acro fau n a in a sandy sedim ent in the
so u th ern N o rth Sea. IC E S Journal o f M arin e Science, 49:
5-11.
B ergm an, M . J. N ., a n d v a n Santbrink, J. W . 2000. M ortality in
m egafaunal b e n th ic p o p u latio n s caused by traw l fisheries on
th e D u tc h co n tin e n ta l shelf in th e N o rth Sea in 1994. ICES
Jo u rn a l o f M a rin e Science, 57: 1321-1331.
B raber, L ., and d e G ro o t, S. J. 1973. T he food o f five flatfish
species (Pleuronectiform es) in the so u th ern N o rth Sea.
N e th erlan d s Jo u rn a l o f Sea Research, 6: 163-172.
Brey, T . 1999. G ro w th perform ance and m o rtality in aquatic
m acro b en th ic in v erteb rates. A dvances in M a rin e Biology, 35:
153-223.
1214
R. Callaway et al.
B row n, J., H ill, A. E., F ern an d , L ., a n d H o rsb u rg h , K . J. 1999.
O bservations o f a seasonal jet-lik e circulation a t th e central
N o rth Sea. E stuarine, C o a stal a n d Shelf Science, 48: 343355.
C allaw ay, R., C otter, J., Jennings, S., a n d L a n c aste r, J. 2002.
M esh-size m atters in epibenthic surveys. Jo u rn a l o f the
M arine Biological A ssociation U .K ., 82: 1—8.
C larke, K . R ., a n d W arw ick, R . M . 1994. C h a n g e in m arine
com m unities: A n a p p ro ach to statistical analysis and
in terpretation. N a tu ral E n v iro n m e n t R esearch C ouncil, U K .
144 pp.
Collie, J. S., E scanero, G. A., a n d V alentine, P. C. 1997. Effects
o f b o tto m fishing on benthic m eg a fau n a o f G eo rg es Bank.
M arin e Ecology Progress Series, 155: 159-172.
C onnell, S. D ., a n d A nderson, M . J. 1999. P re d a tio n by fish on
assem blages o f intertidal epibiota: effects o f p re d a to r size and
patch size. Journal o f E x p erim en tal M a rin e B iology and
Ecology, 241: 15-29.
D ooley, H . D . 1974. H ypothesis concerning th e circu latio n o f
th e N o rth Sea. Journal du C onseil, 36: 54-61.
D uineveld, G. C ., K iinitzert, A. A ., N ie rm a n n , U ., D e W ilde,
P., a n d G ray, J. S. 1991. T he m a c ro b en th o s o f th e N o rth Sea.
N etherland s Journal o f Sea R esearch , 28(1/2): 53-65.
D yer, M . F., Frey, W. G ., F re y , P. D ., a n d C ra n m e r, G . J.
1983. B enthic regions w ithin th e N o r th Sea. J o u rn a l o f the
M arin e B iological A ssociation U .K ., 63: 683-693.
F rau en h eim , K ., N eum ann, V ., T hiel, H ., a n d T ü rk ay , M.
1989. T he distribution o f larger e p ifa u n a du rin g su m m er and
w inter in th e N o rth Sea a n d its su itability fo r environm ental
m onitoring. Senckenbergiana m aritim a, 20(3/4): 101-118.
Freese, L., A uster, P. J., H eifetz, J., a n d W ing, B. L . 1999.
Effects o f traw ling on seafloor h a b ita t a n d a sso ciated inver­
teb ra te ta x a in th e G u lf o f A la sk a . M arin e E cology Progress
Series, 182: 119-126.
F rid , C. L. J., H arw o o d , K. G ., H aii, S. J., a n d H aii, J. A. 2000.
L ong-term changes in the b e n th ic com m unities o n N o rth Sea
fishing g rounds. ICES Jo u rn a l o f M arin e Science, 57(5):
1303-1309.
F rid , C. L. J., a n d Haii, S. J. 1999. Inferring chan g es in N o rth
Sea b en th o s fro m fish sto m ac h analysis. M a rin e Ecology
Progress Series, 184: 183-188.
G lem arec, M . 1973. T he benthic co m m u n ity o f th e E uropean
N o rth A tlan tic continental shelf. O cean o g rap h y a n d M arine
Biology, A n n u a l Review, 11: 263-289.
H a m o n d , R . 1969. On the N o rfo lk m arin e a re a , a n d the
offshore sta tio n s w orked w ithin it. T ra n sa c tio n s N o rfo lk and
N orw . N a tu ra l Society, 21: 209-228.
H ill, M . O. 1973. D iversity a n d evenness: A unifying n o tatio n
a n d its consequences. Ecology, 54: 427-432.
Jennings, S., L ancaster, J., W oolm er, A ., a n d C o tte r, J. 1999.
D istrib u tio n , diversity and a b u n d a n c e o f ep ib e n th ic fa u n a in
th e N o rth Sea. Journal o f th e M a rin e B iological A ssociation
U .K ., 79: 385-399.
Jennings, S., D inm ore, T. A., D u p lisea, D . E., W a rr, K. J., and
L ancaster, J. E. 2001. T raw lin g d istu rb an c e c a n m odify
benthic p ro d u c tio n processes. Jo u rn a l o f A n im al Ecology,
70: 459-475.
K aiser, M . J., R am say, K ., R ic h a rd so n , C. A., Spence, F . E.,
an d B rand, A. R. 2000. C h ro n ic fishing d istu rb a n c e has
changed sh elf sea benthic com m unities. Jo u rn a l o f A nim al
Ecology, 69: 494-503.
K aiser, M . J., A rm strong, P. J., D a re , P. J., a n d F la tt, R. P.
1998. B enthic com m unities associated w ith a heavily fished
scallop g ro u n d in the English C hannel. Jo u rn a l o f th e M arine
Biological A ssociation U .K ., 78(4): 1045-1059.
K n ijn , R . J., B oon, T . W ., H eessen, H . J. L ., a n d H islop,
J. R . G . 1993. A tlas o f N o rth Sea fishes. IC E S Co­
op erativ e R esearch R e p o rt, 194: 1-268.
K rö n c k e, I. 1990. M a c ro fa u n a sta n d in g stock o f th e D ogger
B ank. A com parison: II. 1951-1952 versus 1985-1987. Are
changes in th e co m m u n ity o f the n o rth ea stern p a rt o f the
D o g g er Bank due to environm ental changes? N etherlands
Jo u rn a l o f Sea R esearch, 25: 189-198.
K ü h n e , S., and R a c h o r, E, 1996. T h e m acro fau n a o f a stony
sand area in th e G e rm an B ight (N o rth Sea). H elgoländer
M eeresuntersuchungen, 50: 433-452.
K ü n itzer, A., B asford, D ., C raeym eersch, J. A ., D ew arum ez,
J. M ., D orjes, J., D uineveld, G. C. A ., E leftheriou, A., H eip,
C., H erm an , P., K in g sto n , P., N ie rm a n n , U ., R a ch o r, E.,
R u m o h r, H ., a n d deW ilde, P. A. J. 1992. T h e benthic infauna
o f th e N o rth Sea: species d istribution a n d assem blages. ICES
Jo u rn a l o f M arine Science, 49: 127-143.
L indeboom , H . J., a n d deG ro o t, S. J. (eds) 1998. Im pact-II,
T he effects o f different types o f fisheries on th e N o rth Sea and
Irish Sea b enthic ecosystem s. - N IO Z R a p p o rt 1998-1R IV O -D L O R e p o rt C003/98.
P etersen, C. G . J. 1914. V aluation o f th e sea. II. T he anim al
com m unities o f th e sea b o tto m a n d their im portance for
m arin e zoography. R e p o rt o f the D a n ish Biological Station,
21: 1-44.
P icton, B. E., a n d H ow son, C. M . (eds) 1999. T he species
directo ry o f th e m a rin e fauna a n d flo ra o f th e B ritish Isles
and surro u n d in g Seas. U lster M u seu m P u blication N o. 280.
Rees, H . L ., Pendle, M . A ., W aldock, R ., L im penny, D . S., and
B oyd, S. E. 1999. A com parison o f benthic biodiversity in the
N o rth Sea, E nglish C hannel, a n d C eltic Seas. IC E S Journal
o f M a rin e Science, 56: 228-246.
R ijn sd o rp , A. D ., Buijs, A. M ., S torbeck, F ., and Visser, E.
1998. M icro-scale d istribution o f b e am traw l effort in the
so u th e rn N o rth Sea betw een 1993 a n d 1996 in relation to the
traw lin g frequency o f th e sea bed a n d th e im pact on benthic
organism s. IC E S Jo u rn a l o f M arin e Science, 55: 403^419.
R ogers, S. I., R ijn sd o rp , A. D ., D am m , U ., a n d V anhee, W.
1998. D em ersal fish p o p u latio n s in th e coastal w aters o f the
U K a n d contin en tal N W E urope fro m beam traw l survey
d a ta collected fro m 1990 to 1995. Jo u rn a l o f Sea R esearch,
39: 79-102.
R u m o h r, H ., and K ujaw ski, T . 2000. T he im pact o f traw l
fishery on the e p ifa u n a o f th e so u th e rn N o rth Sea. IC E S
Jo u rn a l of. M arin e Science, 57: 1389-1394.
S im pson, J. H. 1994. In tro d u ctio n to th e N o rth Sea project. In
U n d e rstan d in g th e N o rth Sea system , pp. 1-4. Ed. by H.
C h a rn o ck , K . R. D yer, J. M . H u th n a n ce , P. S. Liss, J. H.
Sim pson, and P. B. T ett. C hapm an a n d H aii fo r th e R oyal
Society, 222 pp.
S p arh o lt, H . 1990, A n estim ate o f th e to ta l biom ass o f fish in
th e N o rth Sea. Jo u rn a l du Conseil, C onseil In tern atio n al
p o u r l’E x p lo ra tio n d e la M er, 46: 200-210.
T urrell, W . R ., H en d erso n , E . W ., Slesser, G ., Payne, R ., and
A d am s, R . D . 1992. Seasonal changes in th e circulation o f
th e n o rth ern N o rth Sea. C o n tin en tal Shelf R esearch, 12:
257-286.
Z ühlke, R ., A lvsvâg, J., de Boois, I., C o tte r, J., F o rd , A.,
H inz, H ., Jarre-T eichm ann, A ., Jennings, S., K röncke, I.,
L ancaster, J., Piet, G ., and Prince, P. 2001. E pibenthic
diversity in th e N o rth Sea. In B urning Issues o f N o rth Sea
E cology, pp. 99-372. E d. by I. K rö n ck e, M. T ü rk ay , and
J. Sünderm ann. Proceedings o f th e 14th Intern atio n al
S enckenberg C onference N o rth Sea 2000, Senckenbergiana
m aritim a, 31(2): 269-281.